A Study of the
SOCIAL AND ECONOMIC
IMPACT OF ODORS
Phase »
Final Report Prepared for the
ENVIRONMENTAL PROTECTION AGENCY
November 1971
COPLEY INTERNATIONAL CORPORATION
-------
A Study of the
SOCIAL AND ECONOMIC IMPACT OF ODORS
Phase II
Prepared for the
ENVIRONMENTAL PROTECTION AGENCY.
Under Contract No. CPA 70-116
By
COPLEY INTERNATIONAL CORPORATION
7817 Herschel Avenue
Lajolla, California 92037
Contributions were made to this report by
POPE, EVANS AND ROBBINS
564 Market Street
San Francisco, California 94104
November 1971
-------
ACKNOWLEDGMENTS
Copley International Corporation gratefully acknowledges the cooperation
of many private citizens of the Los Angeles metropolitan area who gave gener-
ously of their time by participating in personal interviews. Without their whole-
hearted cooperation, this report would not have been possible.
Copley International Corporation is especially grateful to Mr. Richard C .
Dickerson of the Environmental Protection Agency for his counsel and assistance
in the development and conduct of the study, and to Dr. Donald G. Gillette and
Mr. Norman A. Huey of the same Agency for their suggestions relating to the
research.
The work upon which this report is based was performed pursuant to
Contract No . CPA 70-116 with the Environmental Protection Agency.
CONTRIBUTIONS
The overall responsibility of this study was undertaken by Mr. R. David
Flesh, Director, Environmental Economics, Copley International Corporation.
Others who contributed to this report included:
Dr. Amos Turk, Project Consultant and Professor, City College of the
City University of New York.
Messrs. James C. Burns, Senior Staff Chemist; P.M. Cohn, Vice
President; Robert E . Allen and Dante M . Esta, Technical Assistants; of Pope,
Evans and Robbins.
Mrs. Marian O. Doscher, Senior Industrial Economist; Messrs.
Thomas H. Copeland, Director, Marketing and Behavioral Research; Stephen J.
Huxley, Andrew A. Mauro, and R. Paul Weddell, Economists; and Mrs. Linda E
Hanson, Manuscript Typist; of Copley International Corporation.
-i-
-------
TABLE OF CONTENTS
ACKNOWLEDGMENTS AND CONTRIBUTIONS i
LIST OF TABLES vii
LIST OF FIGURES xi
CHAPTER
I INTRODUCTION 1
NATIONAL SURVEY OF THE ODOR PROBLEM .... 1
SOCIAL AND ECONOMIC IMPACT OF ODORS 5
REFERENCES 8
II FIELD PROGRAM PREPARATION 9
SELECTION OF A METROPOLITAN AREA 9
DESCRIPTION OF THE SOURCES 12
SELECTION OF TEST AND CONTROL AREAS 12
NATURAL CONDITIONS ENCOUNTERED 21
REFERENCES 21
IE CONCEPTUAL PROBLEMS WITH PUBLIC ATTITUDE
SURVEYS 23
THE PROBLEM OF BIAS 23
THE PROBLEM OF ESTIMATING TRUE ATTITUDES . 26
REFERENCES 29
IV PUBLIC ATTITUDE SURVEYS 31
DEVELOPMENT OF THE.SURVEY 31
SURVEY FINDINGS 34
SUMMARY 40
REFERENCES 42
-iii-
-------
TABLE OF CONTENTS (Cont'd)
CHAPTER
V TECHNICAL FIELD STUDIES 43
CHOICE OF TECHNIQUES . 43
APPLICATION OF THE TECHNIQUES 47
RESEARCH FINDINGS 54
SUMMARY AND CONCLUSIONS. „ . . 58
REFERENCES 61
VI ECONOMIC EFFECTS OF ODORS 63
SINGLE FAMILY PROPERTY VALUE EFFECTS ..... 64
COMMERCIAL AND INDUSTRIAL EFFECTS 84
OTHER ECONOMIC EFFECTS OF ODORS ....... 86
SUMMARY AND CONCLUSIONS 87
NOTES 88
REFERENCES 89
VII DEVELOPMENT OF ASSESSMENT PROCEDURES 91
ODOR COMPLAINTS AS A BASIS FOR PROBLEM
IDENTIFICATION 92
PROPER BASIS FOR PROBLEM IDENTIFICATION ... 95
POSSIBLE ALTERNATIVE BASIS FOR PROBLEM
IDENTIFICATION 104
POSSIBLE METHODS FOR PROBLEM ASSESSMENT . . 107
SUMMARY 109
NOTES 110
REFERENCES 110
APPENDICES 113
APPENDIX
A FIELD PROGRAM PREPARATION DATA 115
B PUBLIC ATTITUDE SURVEY ADMINISTRATION 125
C PUBLIC ATTITUDE SURVEY FINDINGS 141
-iv-
-------
TABLE OF CONTENTS (Cont'd)
Page
APPENDIX
D TECHNICAL FIELD STUDIES DATA 181
E EFFECTS DETERMINATION QUESTIONNAIRE 193
F GLOSSARY OF TERMS . . 201
ANNOTATED BIBLIOGRAPHY 205
PROCEDURE MANUAL . 253
-v-
-------
LIST OF TABLES
Table
CHAPTER II
Thirty-one potential odor problem areas identified during
the national survey of the odor problem, by U. S. Census
Region 10
List of test areas, control areas, and odor sources
studied in the Los Angeles metropolitan area by month
in which studies were conducted 18
Socioeconomic characteristics of all matched test and
control areas studied in the Los Angeles metropolitan
area (1960 data). 19
Socioeconomic characteristics of all matched test and
control areas studied in the Los Angeles metropolitan
area (1970 data) 20
CHAPTER IV
Number of interviews completed in each test and control
area by month in which public attitude surveys were con-
ducted 35
Cross-tabulations of responses or combination of
responses to questions by respondents' family income,
education, sex, age, and length of residence in Test
Area A 41
CHAPTER V
Concentrations of tertiary dodecyl mercaptan in mineral
oil used to simulate four strengths of refinery odor .... 50
Results of triangle and odor intensity tests in prepara-
tion for the December 1970 field studies 51
-vi i-
-------
LIST OF TABLES (Cont'd)
Table
CHAPTER V (Cont'd)
Results of laboratory correlation obtained by taking scent-
ometer readings of reference standards prepared for an
odor judgment panel 55
CHAPTER VI
10 Single family home sale coverage in Holly Glen (1968-
1970) 70
11 Single family home sale coverage in Southwest Park
(1968-1970) 71
12 Distribution of home loan types in Southwest Los Angeles
County „ 74
13 Average sales price of homes in Holly Glen and Southwest
Park for half year intervals (1965-1970) 78
14 Average sales price of homes in Holly Glen by single
source for half year intervals (1965-1970) 80
15 Average sales price of homes in Southwest Park by
single source for half year intervals (1965-1970) 81
16 Turnover rates of homes per 100 units 84
CHAPTER VII
17 Application of "z" statistics to the results of the public
attitude surveys 98
18 Percentage of respondents who were bothered by odors
and socioeconomic characteristics of control areas de-
termined during the national survey of the odor problem . .
-vin -
-------
LIST OF TABLES (Cont'd)
Table ' Page
CHAPTER VII (Cont'd)
19 Percentage of respondents who noticed and were bothered
by odors and socioeconomic characteristics of control
areas determined during the present study 101
20 Sequential sampling plan for deciding whether or not an
odor problem exists in a community being surveyed.
(AOL =0.263, OPL = 0.410, a = 0.05, & =0.05) 105
21 Percentage of respondents in test areas affected by the
odorous emissions of oil refineries who were bothered
by odors that they noticed 108
-ix-
-------
LIST OF FIGURES
Figure
CHAPTER II
1 Sections of Test Area A and Control Area A studied in
December 1970 (Communities of Hawthorne and Torrance) . 13
2a Sections of Test Area B and Control Area B studied in
December 1970 (Community of Glendale) 14
2b Sections of Test Area B and Control Area B studied in
December 1970 (Community of Torrance) 14
3 Sections of Test Area A and Control Area A studied in
March and June 1971 (Communities of El Segundo,
Hawthorne, and Torrance) 15
4a Sections of Test Area B and Control Area B studied in
March 1971 (Community of Beverly Hills) 16
4b Sections of Test Area B and Control Area B studied in
March 1971 (Community of Torrance) 16
5a Hourly wind direction and velocity recorded at Los
Angeles International Airport, December 1, 1970 22
5b Hourly wind direction and velocity recorded at Los
Angeles International Airport, March 16, 1971 22
5c Hourly wind direction and velocity recorded at Los
Angeles International Airport, June 15, 1971 22
CHAPTER III
Probability functions of a variable (x), representing "noise"
as measured in control groups and "noise" plus "signal" as
measured in test groups 28
-xi-
-------
LIST OF FIGURES (Cont'd)
Figure
CHAPTER V
7 The Scentometer 45
8a Example of calibration test results when no odor was
encountered in Test Area A 53
8b Example of calibration test results when no odor was
followed by sudden odor in Test Area A 53
8c Example of calibration test results when odor was bare-
ly discernible or very transient in Test Area A 53
9a Example of odor intensity readings obtained by mobile
scentometry during morning traverse of Test Area A ... 56
9b Example of odor intensity readings obtained by mobile
scentometry during afternoon traverse of Test Area A ... 56
10 Odor Intensity Rating Sheet . „ „ 57
11 Correlation of scentometer readings and panel ratings
developed during the national survey of the odor problem
(Phase I) and the present study (Phase II) 60
CHAPTER VI
12 Location of real estate developments known as Holly Glen
and Southwest Park 67
13 Mean price of homes in Holly Glen and Southwest Park,
1965 - II = 100 79
CHAPTER VII
14a Radiophone Message Log 93
14b Nuisance Complaint Form 93
-xii-
-------
LIST OF FIGURES (Cont'd)
Figure Page
CHAPTER Vn (Cont'd)
15 Method of odor problem identification and index determi-
nation using the Problem Identification Questionnaire. ... 99
16a Sequential sampling chart indicating an odor problem in
TA(Hawthorne) during the June 1971 public attitude survey . 106
16b Sequential sampling chart indicating no odor problem in
CA(Torrance) during the June 1971 public attitude survey . 106
-xni-
-------
CHAPTER I
INTRODUCTION
Under the Air Quality Act of 1967 and the Clean Air Amendments of 1970,
the federal government embarked on a program to amass information relating to
the sources, effects, measurements, and controls of odors and other acknowl-
eged air pollutants. This comprehensive effort will support the Environmental
Protection Agency in drafting and promulgating a series of air quality, perfor-
mance, and emission standards to be implemented at the state level. For the
solution of countless community odor problems, it is deemed appropriate that
the standards specify what levels of odors are acceptable or how such levels
should be determined, particularly at ambient intensities.
Numerous studies have already been performed under this program, in-
cluding the national survey of the odor problem, conducted in 1969, and the
present study, conducted in 1970 and 1971. Both studies investigated methods
of evaluating community odor problems and, together, yielded information
toward the specifications alluded to above. Both studies were conducted by
Copley International Corporation for the Environmental Protection Agency,
under contracts CPA 22-69-50 and CPA 70-116.
NATIONAL SURVEY OF THE ODOR PROBLEM
Objectives and Methodology
The overall objective of this first phase of research was an assessment
of the national odor problem based on the projections of the results of investi-
gations in a selected sample of identified odor problem areas. Tasks leading
to the achievement of this objective included:
(1) The identification and description of major odorants and
odorant sources.
(2) A determination of the odor producing potential and the
location of groupings of industrial odorant sources .
(3) The location of concentrations of population.
(4) A determination of the effects of the natural environment
on odor problems and the identification of odor supporting
atmospheric regions.
-1-
-------
(5) A survey of all local air pollution control agencies in the
United States.
(6) The selection of seven metropolitan areas for investigations
of possible odor problems .
(7) A series of technical field investigations involving the use
of sensory techniques.
(8) A series of public attitude surveys of residents and busi-
nessmen.
The presence of an odor problem is dependent upon the existence of an
odorant source (or sources), an odor receptor population, and a mechanism to
transport the former to the latter. Thus, in order to ultimately assess the
national odor problem, it was necessary to identify areas throughout the United
States with coincidence of high levels of three basic factors:
(1) Industrial odor producing potential, represented by the
number of operating industrial plants weighted by relative
size of plant and detectability of odorants produced.
(2) Potential extent of odor reception, represented by the size
of the population affected.
(3) Atmospheric vulnerability, represented by a composite of
the major causes of impaired atmospheric dilution capacity.
A sample of seven metropolitan areas was selected from a list of 31 found
to rank high in these factors. The selection process took into consideration the
need for nationwide representation of the study results and detailed information
on likely odor problems solicited from the local air pollution control agencies .
At least one problem community in each of the seven metropolitan areas was
evaluated using scentometers to measure the intensity and areal extent of the
odors detected. In Philadelphia, PA, an odor judgment panel was employed to
determine intensity patterns of odors caused by two sources.
The public attitude surveys were conducted as an exploratory approach
to determine the extent to which odors constitute an influence on the lives of
people living and working in problem communities . The objectives were to
determine if people are generally aware of odors in their communities, if the
public is interested in the abatement of odors, if people believe they have
suffered socially or economically because of odors, and if differences in opin-
ion about odors exist between people living in odor affected (test) areas and those
living in relatively odor free (control) areas.
-2-
-------
Principal Findings of the National Survey
The principal findings of the first phase of research were drawn from the
results of the technical field investigations and the surveys of local air pollution
control agencies and public attitude.
Results of the Survey of Local Air Pollution Control Agencies . The
responses received from the mail survey of 184 local air pollution control agen-
cies in the United States indicated that odor problems existed in many cities and
counties throughout the nation. It was further indicated that the odor problems,
in general, affected a very large number of people.
Four categories of sources were responsible for almost half of the odor
problems reported. These categories were: rendering/meatpacking/slaughter-
houses; chemical/plastic plants; sanitary land fill/dumps/open burning/incinera-
tors; petroleum and natural gas refineries/asphalt plants. In contrast, the four
categories reported to have the highest monthly average number of complaints
were: sanitary land fill/dumps/open burning/incinerators; fisheries and fish
canneries; rubber/tire plants; coffee roasting/spice processing plants. The re-
sponding agencies viewed chemical/plastic plants as the most serious odor sources
and pulp/kraft/wood mills as the sources of the fastest growing problems.
The identification and assessment of community odor problems was accom-
plished by a large number of the reporting agencies using the unaided nose, com-
mon sense, and number of complaints as criteria. An equally large number of
agencies reported that there are many inadequacies in such methods of assessment
and expressed a need for assistance in establishing more effective procedures.
Over 70 percent of the responding agencies took some form of abatement
action. One of the most popular forms of abatement was persuasion. The most
frequently used legislation relating to odor abatement was state or local nuisance
laws. Due to staff limitations at most agencies, as well as the general require-
ment that a considerable number of persons perceive an odor situation as a
nuisance, nuisance laws have been ineffective in abating odors. To the extent
this study could determine, none of the seven metropolitan areas visited had a
record of substantial fine or other judicial penalty levied against an odor law
violator.
Results of the Field Technical Program. The scentometer, a vapor dilu-
tion technique device, was found to provide rapid observations at fixed locations.
From the viewpoint of monitoring low intensity odors, however, the scentometer
seemed to lack the sensitivity of an unaided nose. Using the scentometer in each
of the metropolitan areas visited, refinery installations and chemical facilities
were found to be the most significant sources of odors in terms of areal extent.
-3-
-------
The use of an odor judgment panel provided definitive descriptions of the
areal extent and intensity patterns of odors caused by an oil refinery and an ani-
mal rendering plant. This technique was stated to have substantive application
in those special circumstances where legal liability or compliance questions
require explicit documentation of odorous emissions .
Results of the Public Attitude Surveys . The type of resident found in
both the test and control areas was characterized as lower-middle class. He
tended to be blue collar, middle aged or older, below average in education and
income, and with family size lower than average.
Overall, the difference in attitudes between respondents in the test and
control areas was that relating to air pollution. This difference was reflected
consistently in the volunteered and solicited responses to a variety of questions.
For example, air pollution was rated as a serious or somewhat serious problem
by 65 percent of test area respondents, but only 36 percent of control area re-
spondents .
Residents of both areas appeared most aware of odors in relation to air
pollution. However, of five descriptions of air pollution (smoke or dust, notice-
able odors, irritation of the eyes, nose and throat irritation, haze or fog), the
largest difference of opinion between respondents in the test and control areas
was that relating to noticeable odors. Specifically, "air pollution" meant notice-
able odors to 82 percent of test area respondents, but only 67 percent of control
area respondents. From the surveys, one may infer that odors have been rated
as a serious or somewhat serious problem by about 50 percent of test area re-
spondents, but only 25 percent of control area respondents.
It seemed that the respondents who stated that odors were a problem
were only passively interested in the abatement of odors in their communities .
To some extent, this passivity was due to an obvious lack of knowledge of the
authorities to whom complaints must be directed. The respondents would
approve of additional funds to abate odors in their communities on the condition
that the funds would be supplied by offenders . Beyond this, they seemed resigned
to tolerance of an odorized environment. Homeowners in the test and control
areas who stated odors were a problem felt that odors had reduced the market-
ability of their property.
The owners and managers of a sample of commercial businesses in each
test and control area were also interviewed. In general, the opinions of these
businessmen supported those of the respondents. They were as aware of odors
as the residents, but were less concerned.
-4-
-------
Assessment of the National Odor Problem
The results of the study indicated that, although industrial odor producing
potential exists in most metropolitan areas of the United States, the probability
of encountering community odor problems is greatest in the urban portions
of the Appalachian and Rocky Mountain Regions and the California Coastal Zone.
Projecting the results of the attitude surveys to the nation as a whole suggested
that as many as 25 million residents perceived community odors as problems.
SOCIAL AND ECONOMIC IMPACT OF ODORS
Objectives and Methodology
The ultimate objective of the second phase of research was the develop-
ment of a method to assess the social and economic aspects of community odor
problems . Tasks leading to this end included:
(1) The selection of a metropolitan area in which to conduct
an in-depth study.
(2) A series of public attitude surveys to be conducted at quar-
terly intervals.
(3) A series of technical field studies to be conducted concur-
rently with the public attitude surveys .
(4) A comprehensive analysis of social and economic effects
of odors, particularly as reflected in residential property
values.
(5) The preparation of a set of assessment procedures .
As in the first phase, a metropolitan area was selected from the list of
31 potential odor problem areas. Test areas were established near a variety
of sources to permit a comparison of the effects of odors having vastly different
qualities.
The public attitude surveys and technical field studies were conducted in
December 1970, March 1971, and June 1971. The objective of the public attitude
surveys was to obtain data relative to the influence of odors on the attitudes of
residents. Small sample sizes were used to determine whether the survey
method could be used to obtain quick and decisive answers about the existence
of a variety of possible odor problems.
-5-
-------
The technical field studies were to employ sensory techniques to system-
atically measure the intensity, duration, frequency, and temporal variation of
odors for correlation with the results of the quarterly surveys and to further
compare the operational features of the techniques used in the Philadelphia
investigation.
The assessment procedures were to provide step-by-step instructions for
the evaluation of social and economic effects and to verify the sources and areal
extent of odors detected in any community. Such procedures were prepared with
the intent that they could be used by qualified state and local air quality agency
personnel.
Principal Findings of the Present Study
The principal findings of the second phase of research resulted from the
public attitude surveys, the technical field studies, the economic analyses of
property value differentials, and the development of the assessment procedures.
Results of the Public Attitude Surveys. The residents interviewed were
characterized as middle class. When questioned about odors, they tended to
think only of unpleasant odors. In the test areas and matching control areas set
up to study unpleasant odors, a sizeable proportion of respondents noticed odors;
however, the percentage of test area respondents who reported noticing odors
was significantly larger. A majority of residents who noticed odors were also
bothered by them whether they lived in a test area or a control area. A sizeable
percentage of residents who reported noticing odors felt that odors were a serious
problem.
People were generally unaware of any adverse economic effects of odor
problems, or they did not believe such effects existed. During the winter and
spring surveys, a majority of respondents indicated they had not been forced
indoors by odors, but in the summer more than half of the test area respondents
said this had happened. Generally, people reported that odors had not made
members of the family ill. Definition of the degree to which these problems
had resulted in either adverse social or economic effects was generally beyond
the capability of the respondents .
Results of the Technical Field Studies. Correlation of odor reference
standards to the scentometer illustrated that the scentometer was improperly
characterized as an insensitive device in the results of the national survey.
Although the transiency of the odors encountered may render it insensitive at
fixed stations, it was found quite adequate for discriminating among odor inten-
sities from a mobile platform. Because it can be used to defatigue the operator
when odors are of constant intensity and long duration, the scentometer may be
a more sensitive technique than an odor judgment panel in such situations.
-6-
-------
If odor judgment panel evaluations are to be used as a general means of
enforcing odor regulations, a considerable amount of work is still necessary.
Investigation of an odor problem at short notice may require the availability of
a large number of reference standards. The stability of the standards over
long periods needs investigation. Panel training and logistics require a few days
before a panel can be deployed into the field. Because of the number of people
involved, this becomes a lengthy and expensive procedure. The most efficient
use of such evaluations is not as a tool for routine surveillance, but as a means
for gathering substantial evidence against offenders who will not cooperate with
local authorities. As was concluded in the first phase, this technique is parti-
cularly applicable to those circumstances where legal liability or compliance
questions require explicit documentation of an odorous emission or absolute
verification of an odor source.
Results of the Economic Analysis of Property Value Differentials . The
average price of homes was found to have grown faster in the principal test area
than in its matching control area for the 1968-1970 period when an odor problem
existed in the test area. Three explanations were offered for this finding:
(1) The problem was too transitory to be capitalized into prop-
erty values.
(2) The technique adopted was simply not sensitive enough to
detect the small dollar effects likely in this particular
study.
(3) Some other factor having an opposite effect than odors may
have been present.
Since strong odors were present frequently in the test area beginning in 1969
and since no other factors were found to offset the effects of odors, the most
plausible explanation is the insensitivity of the technique.
Results From the Development of the Assessment Procedures. It was
planned that the national survey of the odor problem and the study of the social
and economic impact of odors would provide a series of relationships that could
be used in the development of federal standards regarding odors . Yet, as this
and other research progressed, it became evident that the measurable reaction
to a vast majority of odor problems was annoyance, not other social or economic
effects. Thus, a general means for dealing with such problems must consider
personal evaluations as central to success . In the application of public nuisance
law, the evaluation of the normal person in the community must be considered.
Public attitude surveys were recommended as the proper means for doing so.
To simplify the requirements for odor problem identification, consider-
ation was given to the elimination of surveys in control areas. The best possi-
-7-
-------
bility of doing so was found by combining the results from the odor free control
areas previously investigated. Preliminary attempts indicated that the same
combined results could be used to reduce the effort required to survey test areas.
This reduction of survey effort seems possible through sequential analysis under
which one of three decisions can be made:
(1) That an odor problem is identified.
(2) That an odor problem does not exist.
(3) That the results are inconclusive and that an additional
interview must be completed.
Copley International Corporation will give further consideration to these possi-
bilities during a third phase of research to be conducted for the Environmental
Protection Agency during 1972.
REFERENCES
Copley International Corporation. National Survey of the Odor Problem. Publi-
cation No. PB-194 376. Springfield, Virginia: U.S. Department of
Commerce, National Technical Information Service, 1970.
Copley International Corporation. National Survey of the Odor Problem: Appendix,
Publication No. PB-194 377. Springfield, Virginia: U.S. Department of
Commerce, National Technical Information Service, 1970.
-------
CHAPTER II
FIELD PROGRAM PREPARATION
A major outcome of the national survey of the odor problem was the iden-
tification of potential odor problem areas throughout the United States. Such
areas were described in terms of the coincidence of high levels of the three
basic factors mentioned in the previous chapter — industrial odor producing
potential, potential extent of odor reception, and atmospheric vulnerability.
The field program of the first phase of research was conducted in the
following metropolitan areas:
• Portland, OR
• Kansas City, MO
• Cincinnati, OH
• Buffalo, NY
• Philadelphia, PA
• Tampa, FL
• San Francisco, CA
These locations were chosen to represent 31 potential odor problem areas iden-
tified in all regions of the United States (Table 1).
The present study was to be conducted in a single metropolitan area as
representative as possible of all those listed in Table 1. The diversification of
industrial source, the socioeconomic characteristics of the population, and the
features of the natural environment were all considered in making this important
choice. The first section of this chapter discusses the selection of Los Angeles,
CA. The second and third sections describe the sources of odors and the delin-
eation of the test areas. The location of comparable control areas is also covered
in the third section. The final section portrays the natural conditions encountered
during the field activities .
SELECTION OF A METROPOLITAN AREA
Selection Criteria
A metropolitan area was to be selected in which at least two communities
were known to be affected by odors of industrial origin. These existing situations
-------
Table 1. Thirty-one potential odor problem areas identified during the nation-
al survey of the odor problem, by U. S. Census Region.
Northeast
Boston, MA
Buffalo, NY
Newark, NJ
New Haven, CT
New York, NY
Philadelphia, PA - Camden, NJ
Pittsburgh, PA
Providence, RI
North Central
Chicago, IL
Cincinnati, OH
Cleveland, OH
Columbus, OH
Dayton, OH
Detroit, MI
Indianapolis, IN
Kansas City, KS - Kansas City, MO
Milwaukee, WI
Minneapolis-St. Paul, MN
St. Louis, MO - East St. Louis, IL
South
Atlanta, GA
Baltimore, MD
Dallas, TX
Houston, TX
New Orleans, LA
Norfolk, VA
Tampa, FL
West
Los Angeles, CA
Portland, OR
San Francisco-Oakland, CA
San Jose, CA
were to meet the following criteria:
(1) One community was to be affected by odors commonly
considered to be unpleasant. Another community was to
be affected by odors usually considered pleasant in mod-
erate strength.
(2) The odors in each community were to be caused by a single
plant or a small complex of similar plants well removed
from other sources.
(3) The odors were to have been a recent occurrence in both
communities, preferably occurring for the first time
within the past two years .
(4) Each of the communities was to contain at least 500 occu-
pied homes.
-10-
-------
The first criterion was chosen to permit a comparison of the social and
economic effects of odors having vastly different qualities, but (ideally) similar
intensities. The second was to benefit the use of sensory techniques for evalu -
ating odor intensities throughout the communities. The third was to provide a
basis for a study of changes in property values over time, which may have been
attributable to odors. The final criterion was simply to ensure that the number
of households sampled for the planned public attitude surveys in each community
would not approach a census of all households in those communities.
Selection of Los Angeles, CA
Copley International Corporation contacted air pollution control agencies
in 21 of the potential odor problem areas listed in Table 1 to locate the needed
communities. (Ten of the areas were avoided because of possible severe and
prolonged winter weather conditions which might have adversely affected the
field work performed during the current project.) Considerable difficulty was
encountered in meeting the third criterion. Either no new industrial plants or
processes had begun operations within the past two years or the emissions of
such plants and processes were tightly controlled in accordance with recently
enacted state and local environmental protection laws.
Visits were made to several metropolitan areas having communities
which were thought to meet the stated criteria. Recent publicity about unpleas -
ant odors drew attention to the Los Angeles metropolitan area. A situation,
seemingly ideal for the present study, was found in the Los Angeles community
of Hawthorne. According to statements in the news media, Hawthorne had been
often exposed to strong hydrogen sulfide, sulfur dioxide, and mercaptan odors
caused by a major oil refinery and chemical plant complex. Complaints about
these unpleasant odors, which had been verified by local authorities, accelerated
from annual rates of less than 20 at the end of 1967 to almost 300 by mid-1970.
Although a community affected by pleasant odors was not immediately evident,
a search for such a situation was directed toward possible sources .
A list of 440 plants involved in the manufacture of foods, beverages,
paints, perfumes, plastics, or soaps in the Los Angeles metropolitan area
was prepared from information compiled during the national survey. These
sources were quickly screened on the basis of size of operation (using number
of employees as a guide), isolation from other sources, and proximity to resi-
dential areas. The remaining plants — about 5 percent of the initial list —
were inspected to determine their suitability for study. Odors caused by two
large bakeries were detected in residential areas of Glendale and Beverly Hills,
and odors from a paint and varnish factory were found in a populated area of
Torrance. In addition, unpleasant odors from a second major oil refinery
were detected when traveling in a Torrance residential area immediately to the
east of that affected by the paint and varnish factory. A difficulty was found in
that none of these sources was newly established. However, because of the
-11-
-------
seemingly ideal situation in Hawthorne and the broad range of odor producing
industries, the tendency of the natural environment to support odor problems,
and the diversity of the population in the Los Angeles metropolitan area, the
third criterion was relaxed.
DESCRIPTION OF THE SOURCES
The oil refinery and chemical plant complex causing odors in Hawthorne
is located in the neighboring community of El Segundo (Figures 1 and 3). The
oldest section of the refinery was built about 60 years ago. The refinery now
employs 1,300 individuals and has a capacity to process 220,000 barrels per
day of crude oil into a variety of intermediate and end products. Most of the
refinery operations are concentrated in the northern part of the compound. Buta-
diene, naptha re-run, cumene, and asphalt plants are situated in the northern
part. Thermal cracking, sulfur dioxide treating, liquid petroleum gas produc-
tion, and alkylation plants are situated just south of these. Aromatics and par-
axylene plants are located in the eastern part of the refinery.
The chemical plant, operated by a different firm, is located just east of
the refinery property limits. The plant recovers certain materials, including
sulfur, from by-products of the refinery operations , Although the chemical
plant has been in operation since 1920, its sulfur production rose sharply after
1967 as a result of new and modernized refinery operations. The plant employs
130 individuals.
The bakery located in Glendale (Figure 2a) began operations in 1927. At
present, it employs 225 individuals. The Beverly Hills bakery (Figure 4a) was
established approximately 40 years ago. It employs about 350 individuals.
Bread is the principal product of both plants .
The paint and varnish factory (Figure 2b) is the newest establishment,
having been built in 1956. Eight hundred employees report to this facility, but
only a fraction of these are involved with the manufacturing processes . The
second oil refinery, like the paint and varnish factory, is located in Torrance
(Figures 2b and 4b). The oldest section of the refinery was built in 1928. The
refinery was modernized in 1966 and now employs 700 individuals and has a
capacity to process 125,000 barrels per day of crude oil.
SELECTION OF TEST AND CONTROL AREAS
Delineation of Test Areas
The odors caused by the two bakeries, the paint and varnish factory,
and the second oil refinery did not seem to be as intense in nearby residential
-12-
-------
El Segundo
o
TA( Hawthorne)*
Oil Refinery
ndale
Chemical Plant
MARINE I AVE.
COMPTC
Manhattan
Beacli
MANHATTAN BEACH
ARTESIA
Hermosa
Beach
G A( Torrance)
King Harbor
Marina
co
Torra
CA(X Torrance
Redondo
-Beach
CARSON
Figure 1. Sections of Test Area
A and Control Area A studied in
December 1970 (Communities of
Hawthorne and Torrance) .
Palos Verdes
Estates
-------
Burbank
TB( Glendale)
Glendale
GUN Q<
x
CB(GlendaleL'
Figure 2a. Sections of Test Area B
and Control Area B studied in December
1970 (Community of Glendale) .
BLVD.
attan
Seacli
OT
mmmm
Paint and Varnish Factory
SCALE
Hermosa
Beach
Figure 2b. Sections of Test Area B
and Control Area B studied in December
1970 (Community of Torrance)
Redondo
•Beacli
Torrance
BLVD.
ICARSON
-------
% Hawthor
% ^
El Segundo
G£
TA( Hawthorne)*
ndale
Chemical 'hint
MARINE I AVE
COM PTC
Manhattan
Beach
MANHATTAN BEACH
ARTESIA
Hermosa
Beach
King Harbor
Marina
Redondo
•Beach
Figure 3. Sections of Test Area A
and Control Area A studied in March
and June 1971 (Communities of El
Segundo, Hawthorne and Torrance.
Palos Verdes
Estates
-------
SANTA MONO
BEVERLY
Beverly
Hills
ood
rnn * U,
TB(Beverly Hi
WILSHIREBLVD.t
o -
CB(Beverly Hills)
OLYMPIC
PICO
BLVD.
WASHINGTON
Figure 4a. Sections of Test Area B
and Control Area B studied in March
971 (Community of Beverly Hills) .
MONICA
at tan
Beach
BLVD.
COMPTON
MARINE! AVE.
MANHATTAN BEACH
EARDJ
CB(Torrance)lfQer
ARTESIA
TB(Torrance)
T'orrance
CAR90N
SCALE
Hermosa
Beach
Figure 4b. Sections of Test Area B
and Control Area B studied in March
1971 (Community of Torrance).
Redondo
•Beach
-------
areas as those detected in Hawthorne. Certainly, the adverse publicity about
bad smells seemed focused on Hawthorne. Thus, it was decided to concentrate
half of the field work on the situation in Hawthorne and to divide the remaining
half among the possible odor problems in the other communities. For conven-
ient reference, as is indicated in Table 2, the residential areas affected by
odors in Hawthorne were designated Test Area A. The populated areas affected
by odors in the other communities were designated Test Area B. Each of the
residential areas ("sections" of Test Area A and Test Area B) approximated
census tracts. The location of these areas is shown in Figures 1 through 4.
Location of Control Areas
The results of the public attitude surveys scheduled for the test areas
were to be compared to the results of identical surveys scheduled for commu-
nities having similar socioeconomic characteristics as the test areas, but hav-
ing no odors. Matching control areas were located, as indicated in Table 2 and
Figures 1 through 4. (The reader may wish to refer to Appendix A to view the
locations of all test and control areas on a single map.) The socioeconomic
characteristics of the matched communities are given in Table 3 in terms of
1960 data prepared by the Bureau of the Census and in Table 4 in terms of 1970
data obtained from the same Bureau and the results of the public attitude surveys
conducted in the areas by Copley International Corporation in December 1970,
March 1971, and June 1971. With the exception of TA(X Hawthorne)/
CA(X Torrance), the socioeconomic characteristics of the control areas were
generally within 10 percent of those of the test areas. Additional characteristics
considered in the selection of the control areas are mentioned in Chapter IV.
One other criterion was established for the selection of CA(Torrance).
This control area was purposely chosen upwind of the second oil refinery (com-
pare Figures 1 and 2b). The intent of this strategy was to eliminate as many
factors as possible other than odors that might have contributed to property value
differentials between CA(Torrance) and TA(Hawthorne). The results of a com-
parison of property values in these areas is the subject of Chapter VI.
Changes in Test and Control Areas
After the December public attitude surveys, it was decided to substitute
TA(E1 Segundo) for TA(X Hawthorne) because odors were more often detected
beyond the northern property limits of the oil refinery than in TA(X Hawthorne).
In addition, CA(South Torrance) was substituted for CA(X Torrance) to more
closely match the socioeconomic characteristics of TA(E1 Segundo).
Since the results of the December surveys indicated no reaction to odors
caused by either the paint and varnish factory or the bakery, TB(X Torrance),
TB(Glendale), and CB(Glendale) were dropped from further consideration. One
more attempt was made to measure reaction to pleasant odors . TB(Beverly Hills)
-17-
-------
Table 2. List of test areas, control areas, and odor sources studied in the
Los Angeles metropolitan area by month in which studies were conducted.
Odor Source
Month
Area
December Test Area A (two sections):
1970 1. TA(Hawthorne)
2. TA(X Hawthorne)
Control Area A (two sections):
1. CA(Torrance)
2. CA(X Torrance)
Test Area B (three sections):
1. TB( Tor ranee)
2. TB(X Torrance)
3. TB(Glendale)
Control Area B (two sections):
1. CB( Tor ranee)
2. CB(Glendale)
March Test Area A (two sections):
1971 1. TA(Hawthorne)
2. TA(E1 Segundo)
Control Area A (two sections):
1. CA(Torrance)
2. CA(South Torrance)
Test Area B (two sections):
1. TB(Torrance)
2. TB(Beverly Hills)
Control Area B (two sections):
1. CB(Torrance)
2. CB(Beverly Hills)
June Test Area A (two sections):
1971 1. TA(Hawthorne)
2. TA(E1 Segundo)
Control Area A (two sections):
1. CA(Torrance)
2. CA(South Torrance)
oil refinery and chemical plant
oil refinery and chemical plant
second oil refinery
paint and varnish factory
bakery
oil refinery and chemical plant
oil refinery and chemical plant
second oil refinery
bakery
oil refinery and chemical plant
oil refinery and chemical plant
"X" before name of community indicates area studied in December 1970 was re-
placed by alternate area in March 1971. TB(X Torrance) was eliminated from
further consideration.
-18-
-------
Table 3. Socioeconomic characteristics of all matched test and control areas studied in the Los Angeles
metropolitan area (1960 data).
Area/Census Tract No .
TA( Hawthorne) /60 23
CA(Torrance)/6505
TA(X Hawthorne)/6022
CA(X Torrance)/6506
TA(E1 Segundo)/6200/6201
CA(South Torrance)/6512
TB(Torrance)/6502
CB(Torrance)/6500
TB(X Torrance)/6503
CB(Torrance)/6500
TB(Glendale)/3015
CB(Glendale)/3017
TB(Beverly Hills)/7008
CB(Beverly Hills)/7009
Median
Family
Income
$8,456
7,926
7,832
8,662
7,834*
7,998
7,867
8,079
7,683
8,079
7,263
6,768
8,191
8,656
Median
Education
12.6 yrs.
12.3
12.2
12.4
12.3*
12.5
12.3
12.3
12.1
12.3
12.1
12.0
12.5
12.6
Homes Built
in the 1950 's
99.4%
100.0
30.6
80.9
51.9*
82.4
94.1
85.6
81.1
85.6
16.1
13.1
15.0
20.8
Median
Home
Value
$19,300
17,400
14,700
21,600
18,450*
16,800
16,200
17,700
16, 200
17,700
15,500
15, 500
31,700
28,700
* Characteristics averaged together.
Source: U.S. Department of Commerce, Bureau of the Census ,
-------
O
I
Table 4. Socioeconomic characteristics of all matched test and control areas studied in the Los Angeles
metropolitan area (1970 data).
Area/Census Tract No .
TA(Hawthorne)/6023
CA(Torrance)/6505
TA(X Hawthorne)/6022
CA(X Torrance)/6506
TA(E1 Segundo)/6200/6201
CA(South Torrance)/6512
TB(Torrance)/6502
CB(Torrance)/6500
TB(X Torrance)/6503
CB(Torrance)/6500
TB(Glendale)/3015
CB(Glendale)/3017
TB(Beverly Hills)/7008
CB(Beverly Hills)/7009
Median
Family
Income' *)
$13,900
12,800
11,500
13,400
12,100
13,800
12,400
11,800
12,800
11,800
8,800
7,900
14,545
15,000+
Median Education
of Respondents' 1)
13.0 yrs.
12.9
12.7
14.7
12.8
14.8
12.7
12.9
12.9
12.9
12.7
12.7
15.0
16.0
Median Age
of Respondents(l)
42.1 yrs.
40.1
45.3
38.5
47.2
40.5
42.9
41.2
37.4
41.2
48.7
53.0
52.5
53.6
Median
Home
Value<2>
$29,580
27,850
23,705
33,990
30,135*
32,905
25,330
27,350
24,325
27,350
23,860
23,765
50,000+
50,000+
* Characteristics averaged together.
Source: (1) c0piey international Corporation (1970 and 1971 data taken from the results of the public attitude surveys),
(2) U.S. Department of Commerce, Bureau of the Census (preliminary data).
-------
and CB(Beverly Hills) were surveyed in March. Although the residents of the
Beverly Hills test area enjoyed higher median family incomes than the residents
of the Glendale test area, they were not more vocal about bakery odors. All of
the sections of Test Area B and Control Area B were eliminated from further
consideration after the March surveys.
NATURAL CONDITIONS ENCOUNTERED
Topographical Features
The surface of almost all of the test and control areas studied was quite
flat. The single exception was TA(E1 Segundo) which had several hills ranging
from about 100 feet to 200 feet above sea level. For reference in reviewing the
results of the technical field studies in Chapter V, topographical maps of
TA(E1 Segundo) and TA(Hawthorne) are included in Appendix A.
Weather Conditions
Except for rains encountered during the first two days of the December
surveys, the main adversity with the weather was occasional moderate to strong
winds which tended to increase the transiency of any odors being detected.
Estimates of both seasonal and hourly variations in winds in Test Area A
are given in Figure 5 for the first days on which surveys were conducted in each
of the three months . Note that the percentage of westerly winds increased as
summer approached. More detailed local climatological data for the months in
which the surveys were conducted are also included in Appendix A.
REFERENCES
U.S. Bureau of the Census. U. S. Censuses of Population and Housing: 1960.
Census Tracts . Final Report PHC(l)-82. Washington, D.C.: U.S.
Government Printing Office, 1962.
National Oceanic and Atmospheric Administration. Local Climatological Data:
Los Angeles International Airport. Washington, D.C.: U.S. Govern-
ment Printing Office, December 1970, March 1971, and June 1971.
-21-
-------
Figure 5a. Hourly wind direction and velocity recorded at Los Angeles Interna-
tional Airport, December 1, 1971.
16
^
I"
& 8
•i-l
8
"a> /i
> 4
-AM £ PM -
CO
- z &
co W w Z pq W
W i w W
~W I 1
-1 1 1 1
z w
3 w
co
W
1
K»
r<
5
H
3
§i
S £
K1
3
»
>• r«
5
K.
f
r>
5 S
^
= ^ .
s ?
\ ,h:
1 ^ -
1 —
-
1 — ' " — "
36 9 12 3 6 9 12
Time
Figure 5b. Hourly wind direction and velocity recorded at Los Angeles Interna-
tional Airport, March 16, 1971.
16
8 -
o
4
0
- AM
^.
s>
K.
K
K
CO 2
— ^ ^ Z la
M ^ ^ ^ Z
— • ^ co Z -
1 Z H ^ i 1
1 1 III 1
K"
>
>•
>•
K
1*
z 1
£ p
O 0
£ ~
^ £ £ PM
> >•
K
" K
>.
S
K
^ S ^2
P- ^ CO
CO —
^ —
CO
I ~
36 9 12 3 6 9 12
Time
Figure 5c. Hourly wind direction and velocity recorded at Los Angeles Interna-
tional Airport, June 15, 1971.
16
W 10
S12
5 -
0
o
4 -
0
- AM PM -
£
t~*
H!
K"
I °
— co W 1
ia 1
1 1
a p
-1 t
a S
w
r/\
^X J
»»
>
S CO —
^ 1 £
PC;
t-*
•
* ^ §~
& •
1° -
36 9 12 3 6 9 12
Time
Source: U.S. Department of Commerce, National Oceanic and Atmospheric
Administration
-22-
-------
CHAPTER III
CONCEPTUAL PROBLEMS WITH PUBLIC ATTITUDE SURVEYS
The purpose of this chapter is to discuss the conceptual problems encoun-
tered in gathering and interpreting data by means of public attitude surveys . The
first section of the chapter serves as a primer on minimizing distortion to the
results of attitude research. In connection with this, specific techniques employed
to minimize the sources of bias in the surveys undertaken in the Los Angeles
metropolitan area during the current project are discussed. The final section
describes a method advanced by signal detection theory for estimating true atti-
tudes from responses about odors under investigation.
THE PROBLEM OF BIAS
Potential Sources of Bias
Bias can be injected into a public attitude survey from one or more of the
following sources:
• Population definition
• Test and control group selection
• Respondent selection
• Survey response generation
• Questionnaire formulation
• Interview completion
Sequentially, the first potential source for bias in a survey lies in the
area of population definition. The population to be studied must be defined in
such a manner that individual members can be identified. The definition can be
geographical, relating to where the population can be found, and demographical,
relating to the vital statistics of that population. Any number of distinguishing
traits can be utilized in properly defining a population. The potential for bias
occurs when information about a certain population is desired, yet the researcher
defines a somewhat different population. Data generated can be relevant only to
the population tested, yet conclusions are often projected to the population about
which information was desired.
Of equal importance to the proper definition of a population is the proper
division of that population into test and control groups. The only difference be-
tween the two groups should be the presence of the variable to be tested in the
test group (noticeable odors, in this study) and the absence of that variable in
-23-
-------
the control group. An error in the selection of the control group and, hence,
the statistics generated from that group, would affect the measurement of differ-
ence between the two groups and introduce a bias in the findings.
Statistical theory is dependent upon a representative sample, where
every member of the population has an equal chance of being included in the sam-
ple. Such a requirement is satisfied by random selection of the sample, which,
if done properly, creates a mirror-image microcosm of the larger population.
If randomness in the sample selection is not maintained, the researcher incurs a
risk that the statistics generated will not be representative of the population
ostensibly under study. Non-randomness normally occurs when an unsystematic
approach to selecting individual members of the population is utilized. For
example, when sample selection is based on factors such as the pleasing person-
ality of a potential respondent, the easy availability of the respondent, or the
motivation of the respondent to participate, a bias resulting from selection of
an unrepresentative sample is likely to occur .
The bias of an unrepresentative sample can be introduced into a study
even when a sample has been randomly selected. This results when a signifi-
cant percentage of the randomly selected sample is unreachable for interview-
ing or refuses to be questioned. In general, as the response rate to a survey
decreases, the potential for this type of bias increases. The reason for this is
that non-respondents have been shown repeatedly to be different from respondents
to a survey. Whether or not these differences are significant in terms of the
specific variables under study depends upon the individual project. The only way
of insuring against the possibility of bias in this case, however, is to increase
the response rate to as high a level as possible.
The questionnaire, itself, is a tremendous potential source of bias in a
survey. The appropriateness of the data collected in a survey is a direct func-
tion of the precision of the questions asked. Questions which are unclear in
meaning will yield answers which are also unclear in meaning. Misinterpreted
questions introduce bias when their answers are analyzed in terms of the intended
meaning of the questions. Improperly worded questions have the capacity of con-
fusing, embarrassing, antagonizing, or leading the respondent, all of which
introduce bias into the data. Questions which unconsciously reflect the attitude
of the question designer can introduce significant bias into the responses. The
sequence in which questions are asked can have a profound effect upon the re-
sponses obtained. The questionnaire should be structured so that earlier ques-
tions do not exert an unforeseen influence on latter questions.
Another source of bias in the questionnaire occurs when the respondent
is asked for information which is not readily at his disposal. Although respon-
dents often will supply answers to such questions, these replies may represent
totally unfounded guesses.
-24-
-------
Interviewers can introduce bias into a survey in several ways. The most
dramatic example of this is deliberate cheating on the part of the interviewer.
All or part of the questionnaire can be filled out by the interviewer without even
contacting the respondent. To the interviewer, it is quicker, cheaper, and eas-
ier. To the researcher, it means invalid data. Interviewers may consciously
or unconsciously create bias in a survey by leading the respondent, making per-
tinent side comments, changing the wording of questions, failing to ask all ques-
tions, using certain inflections when asking questions, improperly recording a
response, or failing to probe sufficiently to uncover the respondent's true feel-
ings about a question.
In the current project, specific effort was undertaken to avoid or signifi-
cantly decrease the chance for bias in each of the foregoing six areas.
Techniques Employed to Minimize Bias
The populations to be studied in the Los Angeles metropolitan area were
representative of middle-class, adult residents . They were specifically defined
in terms of median family income, median education, median housing value, and
type of housing units. It was intended that they would be exposed to specific odors
of industrial origin. Hence, they were chosen from areas in proximity to odorant
emitting industrial plants . Their neighborhoods were designated as "test" areas.
A myriad of geographical and demographical data was assembled to repre-
sent each test area, after which "control" areas were sought which closely
approximated these same identifying traits . The specification of identical char-
acteristics for test and control groups insured the validity of using such groups
in identifying odor-caused variances. The procedures utilized in test and con-
trol area selection are discussed in Chapter II.
Once test and control groups were selected, a complete schedule of indi-
viduals in these groups, who had listed telephones, was obtained through use of
a street address (reverse order) telephone directory. A random selection of
samples was conducted by professional personnel. Interviewers were not per-
mitted to take part in the selection process, but were given names, addresses,
and telephone numbers of the individuals to be contacted.
Possible bias stemming from non-response was almost entirely eliminated
in the project. This was due to an unusually stringent demand that interviewers
attempt to contact the original respondent six times before substituting alter-
nates. Interviewers were further instructed to make callbacks at varying times
of the day to maximize the chances of finding the respondent at home. Non-
response stemming from refusals to cooperate was controlled through the design
of a brief questionnaire form, the avoidance of antagonizing phraseology, and use
of a highly trained and experienced interviewing staff.
-25-
-------
The many sources of bias that can be associated with the questionnaire
were rigorously controlled through the design. The data to be collected were
generally specified by the Environmental Protection Agency. To meet the re -
quirements, questions were taken from the national survey of the odor problem,
recent studies completed in Sweden and the United States, and the experience of
the Copley International Corporation consumer research staff. Several versions
of a questionnaire were constructed, pretested, reviewed, and refined before a
final form was adopted.
Interviewer bias was reduced by employment of only experienced inter -
viewers, detailed written and verbal briefings of interviewers, one-hundred
percent editing of all completed questionnaires for internal consistency, and
verification of respondents' answers on over 10 percent of the completed ques-
tionnaires .
THE PROBLEM OF ESTIMATING TRUE ATTITUDES
The most difficult problem with public attitude surveys is encountered
with the interpretation of results . It is the problem of estimating true attitudes
from the typically varied responses of the population under investigation. To
attain a high level of confidence in the results, the whole population or at least
many individuals in the population must be interviewed. By doing so, the results
would indicate what the population says about the variable under study, but it
remains uncertain whether they would reflect what the population feels about that
variable.
It could be argued that an individual's responses to carefully designed
survey questions reflect his attitude. Then, varied responses in the results of
the survey would simply indicate varied feelings . Yet, despite the amount of
care given to question design or other aspects of a public attitude survey, an
individual's responses may be biased by a myriad of variables other than the
one of concern. For example, an individual may complain about odors in his
neighborhood because he holds some unrelated resentment against the possible
source, because he believes it is popular to complain about such things, because
he has a strong propensity to neurosis, or because he is bothered by odors.
From this, it can be seen that varied responses in the survey results could
obscure true feelings. Although the essence of this problem lies beyond the re-
searcher's control, an adjustment can be made to minimize its weight.
The use of results of an identical survey of a control group is the key to
estimating true attitudes in the population under investigation. The basis for
this and the related adjustment are most easily discussed in terms of a theory
developed for radar communications, viz ., signal detection theory. Two
descriptors — noise and signal — provide part of the basic nomenclature of
this theory. As such, they are defined and employed extensively below.
-26-
-------
Responses based on the interaction of variables other than the one of con-
cern in the research are called "noise." In the present study, "noise" refers to
responses by individuals that they noticed or, most importantly, were bothered
by odors in their neighborhoods when the location of sources or meteorological
conditions were such that odors (other than low intensity background odors) could
not have been detected. "Noise" in this sense may exist to some level in all test
and control areas . Conversely, "signal" refers to responses by individuals that
they noticed or were bothered by odors when odors were present. By definition,
"signal" can exist only in statistics generated within test areas .
The application of signal detection theory to the current project is graph-
ically portrayed in Figure 6. The left-hand probability function, C, represents
the probability of obtaining from control groups a percentage of responses, X,
that odors were noticed or caused bother. In theory, if many control groups
were properly selected and measured in terms of noticing or being bothered by
odors, the percentages of affirmative responses would approach the normal dis-
tribution represented by C . The average level of "noise" that could be anticipated
from such groups would be represented by the mean of that distribution,
Similarly, T represents the theoretical distribution of the percentages of
affirmative responses from many test groups . The mean of that distribution, Xrp
would represent the average level of "noise" plus "signal" that could be antici-
pated in test group data. An adjustment of the difference between the two means
would give the average level of "signal" alone. The advocates of signal detection
theory have adopted a formula for making such an adjustment . It can be applied
to the results of surveys conducted in any set of test and control areas, but with
less confidence than to the means of distributions such as those described above.
The formula is a Bayesian type of correction for chance success:
P'(Sls) =P(S|s) -P(S[n)
100-P(S|n)
where: P'(S| s) = the percentage of "signal" attributable to the
variable under study
P(Si s) = the total percentage of "noise" plus "signal"
measured in the test group
P(S|n) = the percentage of "noise" measured in the
control group
By this formula, the percentage of control area respondents who say they noticed
or were bothered by odors in their neighborhood is subtracted from the percentage
of test area respondents who give the same reply. The resulting answer is related
to the percentage in the control area who answer "correctly," i.e., who say they
did not notice or were not bothered by odors in their neighborhood.
-27-
-------
Figure 6. Probability functions of a variable (x), representing "noise" as measured in control groups
and "noise" plus "signal" as measured in test groups.
to
oo
I
bo
.3
!
o
fr
V V
Ac AT
Percentage of responses that odors were noticed or caused bother
-------
The following chapter presents the results of the public attitude surveys
conducted by Copley International Corporation in the Los Angeles metropolitan
area. Adjustments to these results, using the formula adopted for signal detec-
tion theory, are discussed in Chapter VII.
REFERENCES
Green, David M., and Swets, John A. Signal Detection Theory and Psycho-
physics . New York: John Wiley and Sons, Inc., 1966.
Swets, John A., Tanner, Wilson P., Jr., and Birds all, Theodore G. "Decision
Processes in Perception," Psychological Review, LXVIII, No. 5
(September, 1961), 301-320.
-29-
-------
CHAPTER IV
PUBLIC ATTITUDE SURVEYS
Public attitude surveys were conducted at quarterly intervals — December
1970, March 1971, and June 1971 — during the present study of the Los Angeles
metropolitan area. Telephone surveys rather than personal, face-to-face inter-
views were used. It was felt that telephone interviewing could adequately generate
the data needed with greater efficiency and economy. The objectives of these sur-
veys were:
(1) To gather data regarding the public's attitudes toward
noticeable odors.
(2) To obtain an estimate of the existence and severity of
community odor problems.
(3) To define the type and degree of social and economic
effects odors have on residents and their property.
This chapter is divided into a description of the survey and a presentation of the
results.
DEVELOPMENT OF THE SURVEY
Description of the Questionnaire
The survey questions chosen originated from five major sources:
(1) The national survey of the odor problem.
(2) A public opinion survey developed in connection with a
pilot study in Eureka, California.
(3) Studies recently completed in Sweden.
(4) Insight gathered from six panel discussions held with
male and female residents of Los Angeles .
(5) Suggestions of the Copley International Corporation con-
sumer research staff.
-31-
-------
As indicated below, the questions were designed to cover topics to serve seven
distinct purposes . They were structured in such a way as to minimize bias re-
sulting from overemphasis of a particular topic. A copy of the questionnaire
used in the public attitude surveys is included in Appendix B.
The first questions were introductory in nature to set the context of the
survey in the minds of the respondents. They dealt with the residents' satisfac-
tion with their neighborhood and perception of community problems . These
questions also provided an opportunity for both unsolicited and solicited opinions
regarding the presence and effects of odors in the neighborhood.
Question 6 was designed to mask the objectives of the survey. This multi-
part question concerned the existence of noise in the neighborhood.
Questions 5 and 7 were designed to determine whether residents felt that
an odor problem existed in their neighborhood. Multi-part question 7 also covered
the extent to which residents were bothered by odors, the source of the odors,
and the severity of the problem .
The next group of questions, numbers 8 through 17, were included to
assess possible social and economic effects of odors. These included the effect
of odors on the activities and health of residents and the impact on home values
and salability of their homes. Questions 8, 16, and 17 were designed to deter-
mine the strengths of respondents' attitudes as measured by willingness to take
positive action to abate odor emissions.
The final set of questions, numbers 18 through 24, covered demographic
data. These questions were designed to determine whether the test and control
areas had been properly matched.
Questionnaire Pretest
Nine questionnaires were administered to residents of Los Angeles com-
munities under consideration as possible test and control areas. The purpose
of the pretest, conducted in October 1970, was to determine whether any diffi-
culties existed relating to the wording or structure of the questionnaire. Minor
alterations to the wording of some questions resulted from this effort.
Survey Technique
During each of the three surveys, telephone interviews were conducted
with a random sample of the population. Interviews were undertaken on consec-
utive days with no more than four or less than three days devoted to any one test
or control area. If the first call to a person selected as a part of the sample
did not result in a completed interview, up to five additional calls were required.
After the sixth call, a person who had not been reached was replaced with another
-32-
-------
chosen when the sample was drawn. This callback procedure insured a high inter-
view completion rate from the original sample and thus increased confidence that
the sample was representative of the area under study.
At least 10 percent of all completed interviews were verified by contact-
ing the respondents a second time. In addition, completed questionnaires were
carefully edited for internal consistency before tabulation of survey results was
begun.
Selection of Test and Control Areas
The Los Angeles metropolitan area was chosen as the location of the
study. The primary reasons for its selection were:
(1) The diversity of the population in terms of a variety of
demographics.
(2) The broad range of industries representing potential odor
producing sources.
(3) Its location in a region of atmospheric vulnerability to
serious odor problems.
(4) The availability of areas in the industrial sector known
to be affected by odorants from single stationary sources .
Telephone discussions with representatives of the air pollution control agency
serving the area confirmed the accuracy of these observations .
Test areas suspected of having odor problems related to specific odor
producing plants were selected. Control areas matching these test areas were
chosen on the basis of the following criteria:
(1) Near the test area but not affected by odors.
(2) Access to freeways similar to the test area.
(3) Median family income not different from the test area by
more than 10 percent.
(4) Median value of owner occupied housing units not different
from the test area by more than 10 percent.
(5) Median gross rent of renter occupied housing not different
from the test area by more than 10 percent.
-33-
-------
(6) Average age of housing units (based on year structures
were built) not different from the test area by more than
15 percent.
(7) Average number of units per structure not different from
the test area by more than 15 percent.
(8) Median number of rooms per dwelling unit not different
from the test area by more than 10 percent.
As is shown in Chapter II, these criteria guided selection of control areas that
approximated the test areas except for the presence of odors.
Conduction of the Surveys
Samples were drawn for the test and control areas. The questionnaires
were then administered by professional interviewers in each of the quarterly
surveys conducted. The total number of interviews completed in each test and
control area is given in Table 5. The location of the test and control areas,
as well as their designations, are described in Chapter II.
SURVEY FINDINGS
The results of each quarterly survey were tabulated for each test and
control area. Where appropriate, tests of statistical significance were made
to support inferences drawn from differences between matched area results.
For the most part, the survey findings focus on the social and economic impact
of odors . To present a different perspective, the findings are categorized by
type of odor source investigated and by principal areas surveyed. Detailed re-
sponses to questions asked of respondents in the principal test and control areas
may be obtained from the tables contained in Appendix C of this report.
Effects of Odors By Type of Odor Source Investigated
Before presenting the findings, it is recalled that three types of odor
sources were investigated in connection with the present study:
(1) Two oil refineries and a chemical plant said to cause un-
pleasant odors.
(2) A paint and varnish factory said to cause neutral (neither
pleasant nor unpleasant) odors .
(3) Two bakeries said to cause pleasant odors .
-34-
-------
Table 5. Number of interviews completed in each test and control area by
month in which public attitude surveys were conducted.
No. of Inter -
Month Area views Completed
December
1970
Test Area A (two sections):
1. TA( Hawthorne)
2. TA(X Hawthorne)
Control Area A (two sections):
1. CA(Torrance)
2. CA(X Torrance)
Test Area B (three sections):
1. TB( Torrance)
2. TB(X Torrance)
3. TB(Glendale)
Control Area B (two sections):
1. CB(Torrance)
2. CB(Glendale)
75
75
50
101
54
51
51
110
51
March
1971
Test Area A (two sections):
1. TA(Hawthorne)
2. TA(E1 Segundo)
Control Area A (two sections):
1. CA(Torrance)
2. CA(South Torrance)
Test Area B (two sections):
1. TB( Torrance)
2. TB(Beverly Hills)
Control Area B (two sections):
1. CB( Tor ranee)
2. CB(Beverly Hills)
75
75
50
50
75
50
75
50
June
1971
Test Area A (two sections):
1. TA( Hawthorne)
2. TA(E1 Segundo)
Control Area A (two sections):
1. CA(Torrance)
2. CA(South Torrance)
75
75
75
75
"X" before name of community indicates area studied in December 1970 was re-
placed by alternate area in March 1971. TB(X Torrance) was eliminated from
further consideration.
-35-
-------
Effects of Odors Caused By the Oil Refineries and Chemical Plant. When
asked whether odors had been noticed in the neighborhood during the previous
three months, a sizeable proportion of both test and control area residents re-
plied affirmatively. About two-thirds of test area respondents compared to half
of the control area respondents had noticed odors. In part, the control area
responses may have been the result of background odors generally found in most
parts of the Los Angeles metropolitan area (question 7A).
Only those respondents who had noticed odors were asked about when and
how often they noticed the odors and how long the odors lasted. Responses from
test and control area residents were very similar. Differences in the time of
day odors were noticed meshed closely with meteorological data on wind patterns
(cf. the Local Climatological Data included in Appendix A). Very few respondents
noticed odors everyday and, while variations in responses among once a week,
once a month, and occasionally were found, no discernible pattern emerged.
With regard to the duration of odors, a majority of both test and control area
residents indicated odors were present for more than one hour but less than a
day (questions 7B, 1C, and 7D).
A very large proportion of residents who noticed odors were bothered by
them in both the test and control areas. At least three-fourths were bothered
by odors. In TB(Torrance), during December nearly 98 percent reported being
bothered. A majority of those bothered reported they were bothered either very
much or much, except in Control Area A in December (question 7E).
Generally, those bothered by odors indicated that this occurred very often
or often (question 7F). An overwhelming proportion of respondents bothered by
odors said the strength of the odor bothered them most. The number of times
the odor was noticed and the length of time it lasted was mentioned as most both-
ersome by less than 20 percent of the respondents in both test and control areas
(question 7G).
Most test area residents were able to identify the source of the odorous
emission which was under investigation. As expected, control area residents were
usually unable to do so. An exception was CB(Torrance). A significant percent-
age of all respondents were unable to indicate any origin for the odors they
noticed (question 7H).
In evaluating the seriousness of the odor problem, of those respondents
who noticed odors, only a very small group felt the situation was less serious
than it was a year ago. Slightly more respondents felt that odors were not a
serious problem. The majority felt that the odor problem was continuous and
serious or was increasing. No pattern of responses was noted with regard to
this latter group, either between test and control areas or over the three quar-
terly surveys (question 71).
-36-
-------
In TA(Hawthorne) and CA(Torrance), residents of both areas said that
air pollution meant, among other things, noticeable odors. December survey
results showed about 80 percent of the residents responded in this way, while
in March and June the proportion exceeded 90 percent. Of those who responded
that some component of air pollution was a problem in their area, more than
half of the test area residents identified odors as a problem. Slightly better
than one-third of control area respondents also gave this opinion. Noticeable
odors received more mention in both test and control areas than did other com-
ponents or effects of air pollution (question 5).
Effects of Odors Caused By the Paint and Varnish Factory. This source
of neutral odor emissions was studied only in December. Test area responses
closely paralleled those of the matching control area on most parts of question 7
relating to odors. The exception was in the response to the seriousness of the
odor problem in that area of the city. About 60 percent of test area respondents
indicated the odor problem was more serious than a year ago, while only a little
more than 30 percent of control area respondents held this view. Since meaning-
ful differences between test and control areas were not discovered, no further
investigation of neutral odors was made. Results of other questions from these
test and control areas are excluded from the findings reported in this chapter.
Effects of Odors Caused By the Bakeries. The effects of pleasant versus
unpleasant odors were studied only in December and March. The December test
and control areas were located in Glendale, while the March areas were in
Beverly Hills. Significant differences were found between the responses in areas
near the refinery and chemical plant on the one hand and the bakeries on the other,
About three-fourths of the Glendale test and control area residents indicated that
no odors had been noticed. In Beverly Hills, an even larger proportion of resi-
dents reported that they had not noticed odors . Sample sizes for all other parts
of question 7 were too small to allow an evaluation of the attitudes about odors of
residents of these test and control areas. It was concluded that no additional
surveys of pleasant odors were warranted. Other findings from these test and
control areas are omitted from this chapter.
Effects of Odors By Principal Areas Surveyed
As indicated in Chapter II, certain areas near the oil refineries and
chemical plant sources were chosen to examine the effect of differing relative
locations on survey results. These areas were also selected to investigate the
results that might be obtained from the use of very small samples. Survey
findings from all but two of these areas have been excluded from the results
described below. Only the findings from TA(Hawthorne) and matching
CA(Torrance) are reported. These test and control areas were most closely
matched and were the only areas studied in each of the three quarterly surveys .
-37-
-------
Social and Economic Effect of Odors. In order to determine whether
odors had produced any economic effects, respondents who were home owners
were asked whether they felt that odor pollution had reduced the value of their
homes . Overwhelming percentages of both test and control area respondents
replied that it had not. Consequently, other parts of the question relating to the
degree to which home values had been reduced provided inconclusive data as the
result of small sample sizes (question 9).
Both test and control area residents reported that houses in the neighbor-
hood were sold usually within three months. However, a sizeable proportion of
respondents did not know how long it took to sell a home in their area (question
10).
A hypothetical situation was posed to respondents bothered by odors
where it would be possible to buy odor free air as it is possible to buy bottled
water. They were then asked how much they would be willing to pay for such
air. In both the test and the control area, more than two-thirds of the respon-
dents replied that they did not know how much they would be willing to pay. It
is speculated that the large proportion of such replies was caused by the newness
of the idea of buying odor free air (question 8).
About three-fourths of the residents in both TA(Hawthorne) and CA(Torrance)
reported that they had never considered moving away because of odor pollution
(question 12). Approximately the same proportion indicated that they had not re-
quested some authority or agency to take action concerning either odors or noise
in their area (question 16). However, about half of the respondents said they
would be willing to sign a complaint about these problems. Control area respon-
dents were more willing to sign a complaint concerning noise than were test area
residents. But, a much larger percentage of persons in both areas were willing
to sign a complaint about odors. Only about one-third of the respondents said
they were willing to appear in court to voice a complaint (question 17).
Residents were asked to consider the advantages and disadvantages for
people in their area of the city and to indicate whether having odor producing
industries nearby was good or bad. A sizeable percentage of respondents in
both test and control areas felt this to be bad. However, a significant percent-
age of test area respondents replied that they did not know (question 11).
When asked how bad odors in a community might possibly affect the daily
activities of residents, a variation over time was noted. In December, 65 per-
cent of test area residents indicated that they did not know. In March, 15 per-
cent gave this reply, while in June, only 6 percent did not know. Conversely,
some effect on health was noted by a very small percentage of December respon-
dents, by almost one-fifth during March, and by more than one-third in June. A
similar pattern of test area responses was noted with regard to curtailment of
outdoor activity. No pattern in control area responses emerged (question 13).
-38-
-------
Slightly more than half of the respondents in both TA(Hawthorne) and
CA(Torrance) indicated a strong odor had not forced them indoors in the
December and March surveys. However, in the June survey more than 50 per-
cent of the test area residents indicated they had been forced indoors while only
25 percent of control area respondents gave this reply (question 14). About
three-fourths of the respondents in both the test area and the control area indi-
cated that odors in the air had never made members of the family physically ill
(question 15).
Neighborhood Conditions. Residents were asked to rate their area as a
place to live. More than two -thirds of both test and control area respondents
rated their area as either excellent or good (question 1). However, when asked
whether they could think of any annoying things in the area that neighbors com -
plain about, almost the same proportion replied affirmatively. An exception
occurred in the control area during the June survey when only 48 percent gave a
yes response. Complaints about air pollution or odors and noise were mentioned
most often (question 2). Respondents were then asked to rate urban problems
with regard to the seriousness of the problem in their area. Of the problems
listed, only air pollution was rated as serious by a sizeable proportion of resi-
dents (question 3).
About three-quarters of the respondents indicated that there were no con-
ditions in their neighborhood that would cause them to spend their leisure time
inside their homes. Those saying that such conditions did exist identified odors
most often. Cold weather or wind and noise were also mentioned by some resi-
dents . About 90 percent of all respondents said that there were no neighborhood
conditions that would cause them to spend their leisure time away from home.
Responses listing specific conditions were too few to permit evaluation (question
4).
As was previously mentioned, questions concerning noise were included
in the survey in order to mask the real intent of the research. About 40 percent
of both test and control area respondents said they had noticed noise in their
neighborhoods. Of those noticing noise, a majority indicated that they had been
bothered by it. Traffic and airplanes were frequently mentioned as the cause of
this noise in the test area. Traffic also received a large percentage of the re-
sponses in the control area. In both test and control areas noise was considered
to be a more serious problem than it was a year ago (question 6).
Demographic Data
From questions 18 through 24 a profile of the respondents to the three
quarterly surveys can be drawn. Employed respondents tended to be white
collar workers. Most respondents were between 35 and 64 years of age and had
a high school education or better. The majority of respondents had annual fami-
ly incomes of $10,000 or more.
-39-
-------
Characteristics of Respondents
In order to determine if a relationship existed between the demographic
characteristics of respondents and their reaction to odors, certain cross-tabu-
lations were made. The responses to selected questions regarding odors were
cross-tabulated with several demographic variables. These were respondents'
family income, education, sex, age, and length of residence in Test Area A.
The results of these tabulations are presented in Table 6.
A larger percentage of female respondents indicated they had noticed
odors than did male residents. Fewer people over 50 years of age noticed
odors than did those in other age groups. Also, a smaller percentage of per-
sons having lived in the area for more than 15 years reported noticing odors.
Although the proportion of men and women who said they had been both-
ered by odors was nearly the same, nearly 10 percent more women than men
were bothered very much or much by odors. Fewer respondents who were high
school graduates or less indicated a willingness to appear in court to voice a
complaint than did respondents with some college. Fewer persons over 50 years
old said they would be willing to appear in court than respondents from other age
groups.
SUMMARY
When questioned about odors, people tended to think only of unpleasant
odors. In both test and control areas set up to study unpleasant odors, a size-
able proportion of residents noticed odors; however, the percentage of test area
respondents noticing odors was significantly larger . A majority of residents
who noticed odors were also bothered by them whether they lived in a test area
or a control area, and most test area residents were able to identify the source
of the odors . A sizeable percentage of residents who reported noticing odors
felt that odors were a serious problem.
People were generally unaware of any adverse economic effects of odor
problems, or they did not believe such effects exist. During the winter and
spring a majority of respondents indicated they had not been forced indoors by
odors, but in the summer more than half of the test area respondents said this
had happened. Generally, people reported that odors had not made members of
the family ill.
Using the formula for problem identification contained in the procedure
manual attached to this report, it was found that odor problems did exist in
TB(Torrance) in December and March and in TA(Hawthorne) in June. However,
definition of the degree to which these problems had resulted in either adverse
social or economic effects was generally beyond the capability of the respondents
-40-
-------
Table 6. Cross-tabulations of responses or combination of responses to questions by respondents' family income, education, sex,
age, and length of residence in Test Area A.
Responses or
Combinations of
Responses
Have noticed odors
during the last
three months
Have noticed and
been bothered by
odors
Have noticed and been
bothered very much
or much by odors
Would be willing to
appear in court to
voice complaint
Family Income
Less
than
$9,999
67%
55
48
30
$10,000
to
$14,999
71%
77
64
39
£15,000
and
more
72%
79
69
43
Education
High
school
grad.
or less
67%
72
64
29
Some
college
or
more
71%
80
62
43
Sex
Male
58%
75
56
42
Female
74%
77
67
30
Age
18
to
34
years
68%
77
65
34
35
to
49
years
80%
71
68
45
50
years
or
more
56%
43
53
25
Length of
Residence
5
years
or
less
73%
76
53
37
6
to
15
years
74%
74
82
38
16
years
or
more
54%
81
53
30
-------
REFERENCES
Jonsson, Erland, Deane, Margaret, and Sanders, George. "Community Reac-
tions to Odors from Pulp Mills: A Pilot Study in Eureka, California."
Paper presented at the Conference on Methods for Measuring and Evalu-
ating Odorous Air Pollutants at the Source and in the Ambient Air,
Stockholm, June 1-5, 1970.
. "On Surveys of Annoyance Reactions in Areas Affected By Odours."
Paper presented at the Conference on Methods for Measuring and Evalu-
ating Odorous Air Pollutants at the Source and in the Ambient Air,
Stockholm, June 1-5, 1970.
"On the Formulation of Questions in Medicohygenic Interview Investi-
gations," Acta_Sociologica, VII, No. 3(1964), 193-202.
-42-
-------
CHAPTER V
TECHNICAL FIELD STUDIES
As a part of the national survey of the odor problem, an investigation was
conducted of the utility of sensory techniques for documenting the intensity of
odors. Two such techniques — vapor dilution measurements and odor judgment
panel evaluations — were employed in this undertaking. The results obtained
from both techniques were described and, to the extent of their use in the
Philadelphia metropolitan area, compared. Although informative, the Philadelphia
studies were not conclusive. So that the techniques could be used most appropri-
ately, additional work comparing the operational features as well as the results
was needed.
The present technical field studies were designed to provide such com-
parative information and to systematically measure the intensity, duration, fre-
quency, and temporal variation of odors encountered particularly in Test Area A.
The measurements were to be conducted concurrently with the public attitude
surveys described in Chapter IV and analyzed to determine whether sensory
techniques can be used to predict the existence of community odor problems.
This chapter is devoted to a comparison of the sensory techniques chosen.
The first section describes the choice of techniques. The second and third sec-
tions deal with the application of the techniques chosen and the results obtained.
The possibility of using such techniques as problem predictors is discussed in
Chapter VII.
CHOICE OF TECHNIQUES
Because the present technical field studies were an extension of those
performed during the national survey of the odor problem, only sensory tech-
niques were used for odor analysis . Chemical and instrumental methods were
not included in the national survey for two reasons:
(1) Odors in the ambient air around an industrial facility are
caused by complex substances composed of many different
odorants. Typically, such substances are present in such
low concentrations as to defy chemical or physical analy-
sis .
-43-
-------
(2) Even though modern instrumentation has enabled some sub-
stances to be detected in the parts per million and even the
parts per billion range, it is not known how to establish
odor intensity levels from such analytical information.
In essence, the reaction between odorants and the olfactory receptor cells
of the nose is not yet understood and, consequently, cannot be imitated. The
most direct method available to study odors, therefore, is the human nose . Even
with the disadvantages of the subjective nature of human evaluation and the possi-
bility of fatigue to the senses, sensory measurements are generally most appro-
priate to the study of human response to odors.
Most sensory techniques of odor measurement employ vapor dilution meth-
ods in which the odor is diluted with increasing volumes of odorless air until it
is just discernible. These measurements can be made by collecting samples and
returning them to the laboratory for dilution and testing, or preferably, by per-
forming these tasks at the site of the investigation. Vapor dilution techniques
are usually classed as static or dynamic methods. "Static" methods are those in
which a sample is diluted physically by means such as syringes and then present-
ed to an individual serving as an odor judge. "Dynamic" methods involve the con-
struction of an apparatus which allows a particular sample to be mixed in any
desired proportion with odorless air. By adjusting the flows of the sample gas
and the odorless air, any dilution can be obtained and presented to the judge.
The simplest device of this type is the scentometer, which was developed
by Norman A. Huey and his colleagues in the late 1950's. The scentometer is
illustrated in Figure 7. The two bulb-shaped pieces shown in the figure fit to
the operator's nostrils. Air is drawn into the device and nose as the operator
inhales. When the four orifices (of diameters 1/2", 1/4", 1/8", and 1/16") are
blocked, air is drawn through activated charcoal beds at top and bottom and is
rendered odorless . By opening any one of the orifices, odorous air is drawn
directly into the device by passing the filters, and is mixed with the odorless
air. The four dilution levels thus obtainable are 2, 7, 31, and 170. These
levels maybe stated as scentometer "readings" of 1, 2, 3, and 4, respectively.
Major advantages of the scentometer are that the device is portable, hav-
ing dimensions of 5" x 6" x 2-1/2" and that measurements can be made rapidly
in the field. Versatility is sacrificed, however, as only four dilution levels are
possible.
In contrast, a more sophisticated dynamic olfactometry apparatus is
available to provide a continuous range of dilutions to threshold. Such apparatus,
however, are more costly and, in some cases, not sufficiently portable to be
carried into the field for direct measurements.
-44-
-------
Figure 7. The Scentometer .
,— Air Inlet to Filter (1/2")
cn
I
Measuring
Orifices
Activated Charcoal
Beds
Glass Nose Pieces
-------
To attain desired portability, the scentometer was selected as the dilu-
tion to threshold device for this study. This choice was made despite certain
difficulties anticipated in its operation.
The scentometer was thought to suffer from three major faults:
(1) An apparent insensitivity when compared to the unaided
nose.
(2) The difficulty of sensing threshold levels, as uncertainty
exists over a considerable range of odorant concentrations.
(3) The dilution to threshold values obtained are not a direct
measure of the intensity of the odor before dilution because
of non-linear relationships resulting from saturation of the
activated charcoal.
To avoid such difficulties, a method involving an assessment of odor inten-
sities without dilution was devised by Amos Turk. Reference standards are pre-
pared from an odorant or combination of odorants having an odor which is
characteristic of that produced by the source under investigation. The lowest
standard intensity must be just above threshold, while the highest standard inten-
sity must be above the strength expected to be encountered in the field. The
number of standards in the series may range from 4 to 12, depending upon the
range of odor intensities expected to be encountered and the ability of a human
being to differentiate between adjacent members in the series.
Odor judgment panelists are selected and trained with the reference stan-
dards until they demonstrate a satisfactory ability to differentiate between the
odor intensities. The panelists are then taken to the test area where they relate
the odor intensity they experience in the ambient air to the standards with which
they were trained. With this technique, rapid changes in odor intensity can be
easily detected while odor levels are continuously recorded.
Because of its unique design for a direct assessment of ambient odors,
the Turk method was chosen as one of the sensory techniques for this study.
This technique was also believed to have three major drawbacks:
(1) The need to experiment with several different series of
standards so that the reference chosen will closely match
the odor caused by the industrial operation being investi-
gated .
(2) The lengthy selection and training procedures necessary
before the panelists were deemed qualified to evaluate
odors in the field.
-46-
-------
(3) The possibility of incurring considerable expense in pre-
paring appropriate standards; selecting, training, and de-
ploying panelists; and analyzing results.
The scentometer and reference standard techniques were recommended
for additional study as a result of similar evaluations performed during the na-
tional survey of the odor problem. It was felt that the inclusion of these tech-
niques in the present study would provide an opportunity to establish their
relative utility for the identification and assessment of community odor problems,
The further and more specific experience, however, revealed several hitherto
unknown problems that inhibited comparison of the techniques and demanded
resolution. These problems are described in following sections of this chapter.
APPLICATION OF THE TECHNIQUES
The Scentometer
The scentometer has been described by its developers and used by at
least 12 local air quality agencies . Most of the descriptions have focused on the
physical characteristics of the device and the principles of operation. Emphasis
in this report is given to experiences gained from actual use.
Four scentometers were obtained from the Environmental Protection
Agency and prepared for use in the technical field studies . Before going into
the field, the orifices of each unit were blocked with tape in such a manner that
they could be opened and reblocked repeatedly. Fresh charcoal was installed
and was repeatedly checked to ensure that all odors were adsorbed by the beds .
Before use in the field, each scentometer was shaken gently in a horizontal
position to distribute charcoal evenly over the beds, thereby eliminating the pos-
sibility of channeling. Each unit was also checked for leakage, particularly at
the connection between the scentometer and the nose pieces. This was done by
using each unit in an area of strong odor after blocking the four orifices . No
discernible odor indicated no leakage .
Four representatives of Pope, Evans and Robbins were trained scento-
meter operators. Each operator was tested to have at least average odor sensi-
tivity by triangle tests (similar to those described in the following section) and
full understanding of operation. The following paragraphs characterize the tech-
nique employed in the field by each operator in Test Area A.
Upon arrival at a fixed station, each scentometer operator neutralized
the effects of ambient odors by inhaling odorless air through the charcoal beds
of the scentometer while keeping the four orifices blocked. When this was
-47-
-------
accomplished, the orifices were opened in turn until the operator could identify
the smallest orifice through which the odor could be detected.
Unfortunately, this technique did not produce usable results for this
study. The method may have been successful for more constant intensities of
odor but, due to the transient nature of odors encountered in Test Area A,
almost no scentometer readings were recorded. Huey recommended using the
scentometer in a car. The application of the technique by a mobile operator
was fairly simple and considerably more effective.
An odorized area was tentatively identified by a preliminary drive around
the oil refinery and chemical plant. The operator then relaxed and inhaled odor
free air, a procedure which required driving outside the test area, for the opera-
tor could not relax with the scentometer bulbs pressed against his nostrils.
When the operator was refreshed, the car was again driven through the
odorized area. Throughout this drive the operator breathed through the scento-
meter with one selected orifice open. The route was immediately retraced with
a different orifice unblocked in order to further define the odor intensity. Each
time the route was traversed, the operator called out the points at which negative
readings (when no odor was detected) changed to positive (when odor was detected),
and vice versa. These points were plotted on a map of the area.
It may be appropriate to have four operators and a driver traverse the
area in one car. In this way, with a different orifice open on each scentometer,
an intensity value can be quickly determined. This requires only one trip through
the area, with a return trip to verify the results.
This method, while it had successful results, entailed some problems.
The automobile had to be well ventilated and maneuverable. For quick results,
several people had to be in the car. If only one operator participated, the suc-
cess of the technique depended upon the constancy of the odorous emission.
Although the mobile technique was found to be effective, the reasons for
this were not fully understood. It can only be speculated that a plume of odor
was more easily detected by the mobile scentometer operators because, with
the effect of the relative velocities of the operator and the transient odor plume,
the transition from no odor to appreciable odor was quite abrupt and noticeable.
Also, conceivably, since the mobile operators surveyed a greater area, their
chance of encountering an odor plume was increased. The stationary scento-
meter operators had less chance of encountering a plume of odor. Even if a sta-
tionary operator received a positive result through one orifice, up to a minute
was needed to complete measurements with the remaining orifices. But, in a
constantly changing odor environment, such time was seldom available. Indeed,
in such an environment, it was difficult to feel certain that a positive result was
obtained at all.
-48-
-------
The Odor Judgment Panel
Twenty people were initially recruited for the December 1970 odor judg-
ment panel. All were instructed that they would be tested on the first day and
that, on the basis of this testing, 16 of them would be retained for the balance
of the week. An attempt was made to employ a group similar in socioeconomic
characteristics to the residents of Test Area A. However, because of the short
term nature of the work, those attracted were mostly housewives, about 40 years
of age.
Screening was conducted in accordance with the procedure of Wittes and
Turk, using triangle tests based on odorous food flavoring extracts. The tests
were simple to prepare, administer, and evaluate. While not a sophisticated
screening procedure, the tests were quite adequate for excluding individuals
with particularly insensitive or irrational responses to odors.
The equipment consisted of a supply of 2-ounce paper cups, water, and
food flavoring extracts such as vanilla and wintergreen. The prospective panel
members were subjected to a series of choices in which three paper cups con-
taining solutions were presented to them. Two of the solutions produced iden-
tical odors, while the third produced a different odor. They were required to
choose the paper cup containing the solution with the different odor.
Strongly scented samples perceivable to all or most of the candidates
were used initially; then the concentration of extract used was gradually de-
creased in water toward the level at which 50% to 75% of them could detect the
odor. As the tests were conducted over several hours, breaks were taken at
regular intervals to avoid fatigue. A Barnebey-Cheney, Model ABB, activated
carbon room purifier unit (4,000 cfrn) was used to ensure the absence of ambient
odor in the test room .
After each test, the scores were reviewed to ascertain the percentage
of correct answers. A range of 50% to 75% correct answers was the target.
If the actual percentage was outside this range, the concentration of odorants
was readjusted. After 12 tests, the scores were reviewed in relation to the
number of panelists required for the field studies . At this point, most of the
candidates were accepted or rejected, although borderline candidates were sub-
jected to further tests until a difference of two correct answers between them
was achieved. The more successful were admitted to the panel. This prelim-
inary screening process took a full day.
Triangle tests were again administered to select an additional 5 people
for the March 1971 panel, since only 11 of the December panel members were
available . Similarly, 6 people were recruited for the June 1971 panel. Of the
10 members of the March panel who were available, 6 had participated on the
December panel. Although it was hoped that 16 panelists would participate each
-49-
-------
day, some absenteeism occurred. As a result, the panels varied from day to
day from 13 to 16 members.
After recruiting and screening the panel, training with reference standards
was undertaken. A non-toxic, stable standard was chosen that closely matched the
oil refinery odor frequently encountered in Test Area A. Using various concentra-
tions of this standard, the panel was trained to recognize the strengths of the odor
with which they would be concerned. Tertiary dodecyl mercaptan was chosen as
the standard. It was dissolved in mineral oil to give four solutions of varying
strengths, as indicated in Table 7.
Table 7. Concentrations of tertiary dodecyl mercaptan in mineral oil used to
simulate four strengths of refinery odor.
Rating
1
2
3
4
Concentration in Solvent
(milliliters/liter)
0.125
0.5
2
8
Layman Assessment
slight
moderate
strong
very strong
The same room used to screen the panel was used for training purposes.
The air was again purified during the training sessions. The standards were
presented in capped, 4-ounce, polyethylene bottles with an insert to minimize
the possibility of leakage caused by shaking the bottle.
Panelists were instructed to shake the bottle gently once or twice, remove
the cap, and squeeze the sides once as they held the bottle under the nose. After
a period of memorizing the different strengths and a few breaks to avoid fatigue,
the panelists were invited to test themselves by removing a standard from the
series at random and attempting to judge its intensity. During this time, the
panel instructors circulated about the room and tested all panelists with unmarked
samples. Prior to training, a few sets of standards were prepared on which the
numbers were not clearly shown but were hidden on labels attached to the bottoms
of the bottles. Panelists were asked to sniff the samples and to identify the
strength of the odor.
The results of the December 1970 triangle and odor intensity tests are
shown in Table 8. Individuals who scored higher than 6 on the triangle test
were accepted. Those with a score of 5 or 6 were retested and accepted or
eliminated based on the results of these additional tests . It is recalled that the
odor intensity test consisted of judging a standard in relation to its proper posi-
tion in a series of four strengths. The performance of any panelist who judged
a standard more than one position away from its proper position was considered
unsatisfactory.
-50-
-------
Table 8. Results
December 1970 field
of triangle and odor intensity tests in preparation for the
studies.
Candidate
Name
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
Sex
F
F
F
F
F
M
F
F
F
F
F
F
F
M
F
F
F
F
F
Age
42
38
50
42
45
21
36
35
45
43
46
46
51
22
69
34
33
33
42
Triangle Test
Score
(Pos. 14)
8
11
5
6
9
8
12
13
13
13
8
8
8
6
6
12
11
5
8
Decision
accept
accept
reject
reject
accept
accept
accept
accept
accept
accept
accept
accept
accept
accept*
reject
accept
accept
accept*
accept
Odor Intensity Test
Number of Times
Score More Than
(Pos. 11) 1 Away
6
6
4
7
5
5
5
5
3
4
5
6
6
5
5
4
0
1
1
1
1
2
0
0
0
0
1
1
0
2
2
1
* After retesting.
When the results of the odor intensity tests indicated that the panelists
were ready to go into the field, a briefing on recording procedure was conducted.
Rating sheets were distributed on which panelists were instructed to record the
highest intensity perceived during each minute of an evaluation.
The quality of the odor perceived was a constant problem. It was impor-
tant that the odor judged was from the oil refinery. The instructions given to
panelists concerning this were explicit. If the panelist was not sure whether
the odor was from the refinery or not, he was to record the odor intensity and
make a notation regarding odor quality. If the panelist was certain that the odor
was foreign, i.e., not of refinery origin, he was instructed to record zero, as
if no odor had been perceived. In all such cases, however, he was to describe
the quality of the odor perceived.
Prior to panel selection and training, both sections of Test Area A were
inspected, and suitable places to station panelists were selected. The locations
were chosen in a systematic manner, so that panelists were approximately
-51-
-------
equally spaced, although possible stations near interfering odor sources, such
as heavily traveled roadways, were avoided.
In addition to selecting 16 stations in each area, convenient places to
perform panel calibration tests had to be found. The test area in this case was
virtually ideal, for parks and a golf course were in proximity downwind. The
parking lots of these places were used as calibration sites.
Panel calibrations were for the purpose of determining the variation be-
tween panelists. All panelists were placed at one station and instructed to record
odor values independently for the same period of time. Independent observations
were achieved by maintaining a distance of at least two yards between individuals.
There are no correct responses to calibration tests . Rather, they serve
to identify variations among panelists' responses. As mentioned above, all
panelists were placed at one station and instructed to record odor values for the
same period of time. In theory, if a panelist regularly records data significant-
ly different from the mean of the group, question arises concerning the accuracy
of the dissenting panelist.
In practice, calibrations did not always work this way. Generally, when
there was no odor, panelists agreed and consistently recorded zeros, as in
Figure 8a. When, after a period of no odor and zero recordings, a noticeable
odor suddenly appeared, this odor was generally perceived and recorded by all
panelists. Figure 8b gives an example of this phenomenon.
Panelists were most varied in their results during calibrations in which
odor presence and strength were questionable throughout. While panelists agreed
on no odor and agreed on no odor followed by sudden odor, barely discernible
odors upset their consistency. Some panelists believed no odor was present,
some perceived slight or strong odors, and any variation in a vague, contestable
odor provoked widely varied changes in intensity ratings . Figure 8c gives an
example of this condition. (Note that from 2:28 p.m. to 2:29 p.m., panelist E
recorded a change from 1 to 3, H recorded a change from 3 to 0, while K record-
ed no odor at all.)
After training the panel with the reference standards and testing the uni-
formity of their responses to actual oil refinery odors, the panel members were
placed at their assigned stations throughout the test area.
To get 16 people to act as a unit and thereby avoid delays which reduce
field evaluation time, a rigid adherence to schedule was required. It was ob-
served during this study that about one hour was the optimum time to leave a
panelist at work. The panelists' concentration, after working in solitude for an
hour, seemed impaired. Therefore, a schedule was used similar to that recom-
mended in Table 3 of the procedure manual attached to this report. When cali-
-52-
-------
Figure 8a. Example of calibration test results when no odor was encountered in Test
Area A.
Time
10:10 a.m.
10:11 a.m.
10:12 a.m.
10:13 a.m.
10:14 a.m.
10:15 a.m.
Panelist
A
0
0
0
0
0
0
B
0
0
0
0
0
0
E
0
0
0
0
0
0
F
0
0
0
0
0
0
G
0
0
0
0
0
0
H
0
0
0
0
0
0
I
0
0
0
0
0
0
J
0
0
0
0
0
0
K
0
0
0
0
0
0
L
0
0
0
0
0
0
M
0
0
0
0
0
0
N
0
0
0
0
0
0
p
0
0
0
0
0
0
Q
0
0
0
0
0
0
R
0
0
0
0
0
0
s
0
0
0
0
0
0
Figure 8b. Example of calibration test results when no odor was followed by sudden odor in
Test Area A.
Time
10:35 a.m.
10:36 a.m.
10:37 a.m.
10:38 a.m.
10:39 a.m.
10:40 a.m.
Panelist
A
0
0
0
0
1
0
B
0
0
0
0
0
2
E
0
1
0
0
1
2
F
1
0
0
0
1
2
G
1
0
0
1
1
2
H
0
1
0
0
1
2
I
0
0
0
0
0
2
J
0
0
1
1
2
2
K
0
0
0
2
2
1
L
0
1
1
1
2
2
M
0
0
1
0
0
0
N
0
0
0
1
0
2
P
0
0
0
0
2
2
Q
0
0
0
0
2
2
R
0
0
0
0
1
2
S
0
0
1
2
0
2
Figure 8c. Example of calibration test results when odor was barely discernible or very
transient in Test Area A.
Time
2:25 p.m.
2:26 p.m.
2:27 p.m.
2:28 p.m. "
2:29 p.m. "
2:30 p.m. "
Panelist
A
1
1
2
1
1
1
B
0
0
2
0
0
0
E
1
1
2
1
3
2
F
1
1
2
2
1
0
G
1
1
3
1
0
0
H
0
1
2
3
0
0
I
1
1
2
3
1
1
J
2
2
2
2
2
3
K
0
2
2
0
0
1
L
1
2
3
2
3
2
M
1
0
0
1
0
1
N
0
1
2
1
1
0
P
0
1
3
0
0
1
Q
1
1
2
3
1
1
R
1
1
2
1
0
1
S
0
2
2
1
0
0
-53-
-------
bration tests were thought necessary, they were included by reducing evaluation
periods by 10 minutes.
RESEARCH FINDINGS
Scentometer Results
During the December 1970 field studies, scentometer measurements were
taken using the stationary method previously described. The operators circulated
among the odor panelists, taking scentometer measurements while the panelists
recorded intensities based on their memory of the reference standards . Using
this method, the operator consistently returned a series of zeros, despite odor
judgment panel ratings of 1 and 2, i.e., "slight" or "moderate." These results
were in accord with the results of the national survey of the odor problem and
findings by other researchers.
Between the December 1970 and March 1971 field studies, laboratory
experiments were also performed to investigate this phenomenon. It seemed
likely that odors which were perceived in moderate strengths by panelists should
be measurable using the scentometer with the largest orifice open. It is recalled
that such a setting represents a dilution level of only 2 .
A laboratory correlation between the scentometer and the odor panel was
performed by the following steps:
(1) One scentometer operator stood ready to take a reading
with a selected orifice unblocked.
(2) Another operator took one of four reference standards,
shook it, removed the cap, placed the bottle under the
open scentometer orifice, and squeezed, telling the scent-
ometer operator to inhale at that time.
(3) The test was repeated using different orifices and differ-
ent standards from the same series (tertiary dodecyl
mercaptan).
The results obtained, using three of the four scentometer operators employed
in this study, are given in Table 9.
As can be seen from these findings, scentometer results should be higher
than those recorded by the odor panelists. For example, when the intensity of
an odor is rated as 1 by odor panelists, a reading of 2 should be obtained with a
scentometer. However, in field use, when such a rating was given by odor panel-
ists , no readings (zeros) were obtained with the scentometer.
-54-
-------
Table 9. Results of laboratory correlation obtained by taking scentometer
readings of reference standards prepared for an odor judgment panel.
Odor Reference Orifice Open (Scentometer Reading)
Standard
1
2
3
4
+ indicates odor was detected; - indicates odor was not detected.
This apparent anomaly was investigated in detail during the March field
studies. It became evident that the reason for the lack of positive scentometer
readings was the transient nature of the odors encountered. Typically, an odor
would traverse a particular site within a few seconds. During that time, it may
have been sufficiently intense for an odor panelist to perceive and evaluate, but
the scentometer operator was left in a state of confusion, unable to be certain of
a positive reading and, thus, forced to record zero .
Using the mobile technique recommended by Huey, the results were con-
sistent with those obtained with the laboratory correlation. To test the success
of the mobile technique, a car containing two scentometer operators, two odor
panelists, and a driver was used. During each trip through an odorized area,
the scentometer operators recorded positive or negative results based on their
scentometer settings, while the panelists recorded values judged against their
memories of the reference standard scale.
In the June 1971 field studies, the mobile technique was used exclusively.
Reliable measurements for particular areas at particular times were obtained.
However, because the odors were so transient, it was difficult to get an accurate
picture of conditions throughout Test Area A over long periods of time. An ex-
ample of area surveillance by mobile scentometry is shown in Figure 9.
Odor Judgment Panel Results
A typical odor panel result sheet is shown in Figure 10. As a rule,
panelists made few, if any, comments . In cases where comments were noted,
they were generally of the type included in the figure.
Panelists returned two to four result sheets per day in the field. All
results were tabulated and analyzed by the panel administrators. Two figures
emerged for each station and evaluation period: (1) the percentage of time odor
was detected, and (2) the mean intensity of that odor. These figures were sum-
marized for each of the technical field studies. Summary tables of the December,
March, and June studies in each section of Test Area A are provided in Appendix
-55-
-------
01
ON
Figure 9a. Example of odor intensity readings obtained by mobile scentometry during morning
traverse of Test Area A.
•
CO
bo
Pi
a
Ui
^n
^V
4->
CO
1
o
.a
rf
r SH
/*\
UJ
CO
3
p
£
13
B
W
E
1
Franklin
ISe
sftind
3
.
4-J
CO
a
4}
0
K4
Oil Refinery
|
Ave.
I
live
,.
55
oJ
•3
CD
z
•
CO
oS
CO
1
2 J
^
.
£
3
PQ
0)
a
CO
0
0
1,000'
-------
Figure 10.
NAME
ODOR INTENSITY RATING SHEET
XQ,*-, DATE &~*
LOCATION
LOCATION
TIME
RATING
COMMENTS
TIME
RATING
COMMENTS
44
4.1
A 9
+S
4*
.5"
•f/
£2.
.r
.5"
f
4*1:
/e>:oo
tt
.f
•ff
09
n
.5*
/a
02.
OJ
IL
tl:of
/t
oL.
.ff
/
67
Of
09
Note: .5 indicates that odor was detected, but was not of sufficient intensity
to be rated as 1.
-57-
-------
D. Maps depicting mean odor intensities and percentage of time odor was detect-
ed at each March and June panelist location are also provided in Appendix D. It
is noted that the mean odor intensity and percentage figures are 4-day averages.
SUMMARY AND CONCLUSIONS
The main purpose of the technical field studies was to monitor malodors
in Test Area A for subsequent use in determining relationships between odor in-
tensity and community reaction. During the early part of the field studies, it
became apparent that modifications to the measurement techniques described in
the national survey of the odor problem were necessary to obtain accurate odor
intensity data. Using these modifications, the reasons for the seemingly poor
correlation between the scentometer and odor judgment panel results reported in
the national survey were identified. Subsequently, a more realistic laboratory
correlation was obtained and verified in the field.
When odors were transient, it was essential to use the scentometer from
a mobile platform. Because of the rapidly changing conditions, a stationary
operator was often unsure whether or not he had obtained a positive reading.
Consequently, a great number of zero findings were recorded at fixed stations.
In contrast, the mobile measurements were successful. This required measur-
ing intensities along the perimeter of the odor source under investigation.
The technicians employed as scentometer operators were known to have
satisfactory triangle test results. During the laboratory and field tests, when
two operators took readings simultaneously, invariably the same result was ob-
tained as long as both operators were certain odor was discernible. Some differ-
ences did occur when the dilution brought odor intensities down to barely
discernible levels (i.e., when one operator was not sure if odor was present and
the other was unable to detect it). Positive results were recorded only when the
observers were certain they could detect odor, and because of this practice,
differences between operators performing measurements under the same test
conditions were rare.
The odor judgment panelists were selected on the basis of triangle test
scores and then trained with selected odor reference standards for a day before
going into the field. The training improved panelists' abilities to discriminate
between odor intensities in the laboratory, but during calibration tests in the
field, some panelists gave results that varied greatly from the mean. This
suggested that the initial screening and subsequent training may have been in-
adequate. A longer selection and training process, including field calibration
tests, might have produced more uniform responses. Application of correlation
factors, based on calibration test results, to normalize panelists' scores was
not feasible because of the limited data and widely fluctuating field conditions.
-58-
-------
A major advantage of the odor panel technique is that readings are made
simultaneously throughout the area under investigation and that a definite pattern
of odor intensity distribution over the area can thus be obtained. This is only
true, however, if the distance between panelists is not too great. During the
national survey, one panelist was at each street intersection throughout the odor
affected area. This arrangement was optimal, but few panelists were needed
since the areal extent of the odors was small.
In the present study, the area was large, while only 16 panelists were
budgeted for use. Two possibilities existed — either to move the panel so that
the area was divided into a series of small areas with panelists on each corner
or to place the 16 panelists throughout the area. The latter procedure was
necessary in this study because of the nature of the data required for odor prob-
lem identification, but the large distances between stations did not provide suf-
ficient data to plot a definite pattern of odor intensities.
If the odor reference standard technique is to be used as a general means
of enforcing odor regulations, a considerable amount of work is still necessary.
Investigation of an odor problem at short notice may require the availability of
a large number of reference standards. The stability of the standards over long
periods needs investigation. Panel training and logistics require a few days be-
fore a panel can be operational. Because of the number of people involved, this
becomes a lengthy and expensive procedure. The most efficient use of the odor
reference standard technique is not as a tool for routine surveillance, but as a
means for gathering substantial evidence against offenders who will not cooperate
with local authorities. This technique is particularly applicable to those circum-
stances where legal liability or compliance questions require explicit documen-
tation of an odor emission or absolute verification of an odor source.
Correlations between odor panel and scentometer readings are illustrated
in Figure 11. It should be noted that the relationship developed during the nation-
al survey of the odor problem was based on actual field measurements of odor
intensities by the two techniques, at fixed locations, whereas the relationship
provided by the present study illustrates the laboratory correlation of odor refer-
ence standards to the scentometer (see Table 9 above). The latter relationship
was verified in the field under transient odor conditions when the scentometer
was used from a mobile platform.
The intent of the laboratory exercise was to illustrate that the scento-
meter was improperly characterized as an insensitive device in the report on
the results of the national survey. Although the transiency of the odors encoun-
tered may render it insensitive at fixed stations, it was found quite adequate for
discriminating among odor intensities from a mobile platform. Because it can
be used to defatigue the operator when odors are of constant intensity and long
duration, the scentometer may be a more sensitive technique than an odor judg-
ment panel in such situations.
-59-
-------
S
I
s
•g
0)
oci
a
o
I
o
Figure 11. Correlation of scentometer readings and panel ratings developed during the national
survey of the odor problem (Phase I) and the present study (Phase II).
1 (slight)
1 (moderate)
Odor Judgment Panel Rating
3 (strong)
-------
REFERENCES
Burns, James C., and Huey, Norman A. Personal communications. March
1971.
Huey, Norman A., Broering, Louis C., Jutze, George A., andGurber, Charles
W. "Objective Odor Pollution Control Investigations," Journal of the Air
Pollution Control Association, X, No. 6 (December, 1960), 441-446.
Wittes, Janet, and Turk, Amos. "The Selection of Judges for Odor Discrimina-
tion Panels," Correlation of Subjective -Objective Methods in the Study of
Odors and Taste. Special Technical Publication No. 440, American
Society for Testing and Materials, 1968.
-61-
-------
CHAPTER VI
ECONOMIC EFFECTS OF ODORS
This chapter describes an investigation of the economic costs attributable
to the presence of industrial odors in a community. Principally, the discussion
is confined to a comparison of single family property values in a test and control
area. Two brief sections comprise the remainder of the chapter. The first of
these discusses the economic consequences of an odorized environment on indus-
trial property and activities. The final section lists other economic effects of
odors and the difficulty that must be overcome for such effects to be properly
assessed.
Discussion of single family property values is centered on an evaluation
of various sources of real estate sales data and an analysis of sales trends in
both TA (Hawthorne) and CA (Torrance). For each source, the accuracy, cover-
age, and problem of obtaining the data are documented. Analysis of the data by
the following operations is fully explained:
• Calculation of average sales prices by half year for the period
1965 through 1970.
• Development and application of deletion rules for investor pur-
chases and extreme values.
• Comparison of single source average sales prices with combined
sources figures.
• Determination of average home value growth rates for each area
and the significance of the differences found.
• Comparison of turnover rates for each area.
As a conclusion to this section, various hypotheses are critically evaluated in
light of the data compiled.
The results of interviews with representatives of major firms in the test
area are offered as implications of the economic consequences of odors on indus-
trial firms. Included among these interviews was one with a manager of a large
industrial park immediately east of the principal odor source. A diversity of
other economic effects and the need for data to substantiate such possibilities
are also mentioned.
-63-
-------
SINGLE FAMILY PROPERTY VALUE EFFECTS
While a review of past studies concerned with air pollution effects on
property values is not fully warranted here, a cursory examination is appropri-
ate since most of the assessment techniques used in these studies were considered
in the design of the present project. To date, all studies done on the relationship
between air pollution and property values have considered air pollution in terms
of dustfall, sulfation rates, and suspended particle concentrations, but not speci-
fically odors. Yet, at the outset of this project, there appeared to be no reason
why this variable could not be added, if only sufficient data were available.
The basic premise in most air pollution/real estate studies is that "...
in a competitive properly functioning completely interdependent general equili-
brium system, all economic losses due to air pollution are capitalized into prop-
erty values. It is further generally assumed that buyers regard land and
structure as a bundle, not as separate entities. Finally, "for pollution to result
in differential property values, buyers need only know that they prefer some
properties to others and, other things being equal, are willing to pay more for
the preferred properties. "2 These primary assumptions plus other theoretical
considerations were most completely set forth by Crocker in his 1970 study of
Chicago property values using FHA data.
By far the favorite technique applied to economic effects research is
multiple regression. Ridker and Henning used this technique in a cross-section
study of St. Louis. Property value and other demographic data were drawn from
the 1960 Censuses of Population and Housing and used on a census tract basis.
Pollution data were drawn from air quality agency sampling stations and interpo-
lated to obtain values for census tracts without stations. Crocker and Anderson,
in their 1969 study, eliminated most of the specification problems of the Ridker
and Henning study and reworked the St. Louis data through an improved model.
They also applied Kansas City and Washington, D.C., data to their model with
good results.
Crocker's 1970 comprehensive study of Chicago embodied another major
advance — viz., switching from ex ante offer price data to ex post sales data.
In addition, he was able to avoid making the heroic assumption that response
error to the Census question on property value was both random and, on average,
offsetting. Evidence on the validity of this point was skimpy, as Crocker noted.
Detailed information collected on two census tracts for the Chicago study did not
offer much support for this common contention.
Another technique for assessing the impact of air pollution on property
value is through the use of professional appraiser judgments. While this method,
employed by Stradbeck, is undoubtedly less sensitive than regression analysis,
it should yield differences if they exist and are large enough.
-64-
-------
As the focus of this study was on odor problems rather than on air pollu-
tion in general, modifications in the approaches of previous studies had to be
made. First of all, a cross-section technique was judged to be unfeasible. It
required the existence of an odor problem over a sizeable area for which various
intensities of odors could be measured. In reality, most odor problems of a
continuing nature exist only over a relatively small area (usually over 1 to 5
census tracts).
A time series regression analysis over a test and control area initially
appeared to be feasible. Adequate property value information seemed to be
available, and both analytical and complaint data from the air pollution control
agency serving Los Angeles County could have been used as a proxy for odor
intensity in the test area. However, the research team was denied access to
the necessary air pollution data and, consequently, was unable to employ this
technique.
If the presence of odors in an area does have a negative impact on home
values, the impact should manifest itself in slower growth in home prices, at
least at the outset of the problem. Thus, an examination of property value growth
rates provides another method of assessment — the one adopted for purposes of
this project. This method could be applied in two ways. First, if a control area
that is homogeneous with the test area, except for odors, could be located, then
the growth in home values could be calculated for each area during a particular
period and compared to isolate the influence of odors. Secondly, if enough data
on multiple sales (the same property sold several times) were available in each
area, then the difference in average change in value could be determined. This
tack necessitates a number of homes being bought at about the same period of
time and held for roughly the same number of months . While sufficient multiple
sales might not be found, compilation of the data would aid in an analysis of turn-
over rates.
Despite the difficulties involved in a growth rate approach, it was hoped
that using a good control area would obviate some of the usual problems. In
addition, it would employ accurate ex post sales data similar to that used by
Crocker in the Chicago study.^ Another consideration was that with good sales
data for the first half of 1970, a spot check could be made on the accuracy of
1970 census tract home value figures. Finally, collecting individual property
sales data over time from several sources would permit a rather complete
assessment of the accuracy and coverage of each source.
Description and Evaluation of Data Sources
Of the several pairs of test and control areas examined in this project,
the growth rate analysis has been limited to TA(Hawthorne), a residential de-
velopment known as Holly Glen (C .T. 6023 — incorporated portion only), and
CA(Torrance), known as Southwest Park (C.T. 6505 — except extreme western
end).
-65-
-------
Holly Glen and Southwest Park are both products of single developers, and
are typical of the better grade housing developments constructed during the mid-
1950's. Both developments offered homes in three floor plans with conventional
concrete foundations, stucco exteriors, composition shingle roofs, three bed-
rooms, and two-car garages, all on 5,000 square foot lots. Initially, the Holly
Glen homes contained between 1,100 and 1,150 square feet and had 1-3/4 baths4,
whereas the Southwest Park homes had the same square footage, but only one
bath. In addition, the Holly Glen homes featured built-in ranges and ovens,
whereas the Southwest Park homes did not. Real estate salesmen operating in
Southwest Park felt it was a quality area. However, several admitted the lack
of a second bathroom lowered sales prices by "several thousand dollars." This
feature undoubtedly explains most, if not all, of the approximately $3,000 or
11 percent price difference between the two developments.
Residents in both areas seemed to be pleased with the quality of the local
schools . A quality education at low cost is possible in both areas, mainly be-
cause of extensive industrial taxable property located within each area's school
district boundaries. One of the major selling points of both areas is their loca-
tion in relatively smog free portions of the Los Angeles basin.
As illustrated in Figure 12, all of the Holly Glen development is contained
within a single census tract and lies entirely within one corporate city limit.
The same is true of Southwest Park, except this development extends a few blocks
across Del Amo Boulevard to the south of the census tract. Because of differ-
ences in the demographics of this extension, the blocks were not included in this
study. During 1969, a new development was constructed on vacant land west of
the north-south portion of Norton Street. Because of the dissimilarity of these
new homes to the majority of homes located in Southwest Park, they were deleted
from the control area.
Real Estate Office Data
Initially, an attempt was made to acquire home sales data for both the
test and control areas by contacting individual real estate firms. A complete
census of all homes for sale in January, 1971 was conducted in both TA(H) and
CA(T) to determine which real estate offices were actively engaged in selling
property in these areas. In Holly Glen, 13 homes were for sale by 9 real estate
dealers. Of these, only two had more than one home for sale in the area. In
Southwest Park, 29 homes were for sale by 17 real estate firms. Here, the
market was a little more concentrated, as 4 companies had 14 homes on the
market. Unfortunately, no realtor dealing in TA(H) was also selling in CA(T).
One realtor in Holly Glen explained that established offices generally confine
their activity to within a radius of two miles. Costs of business licenses as
well as the need for real estate board memberships in several cities were
probably confining factors.
-66-
-------
m 1
El Se^undo.
m % |AVE §
>
< cz
EL S
•*™1^^1™ «^» ^"«
La\vndaleD,
. KQ
MARINE
MANHATTAN
^^•"^•••^••B
AVE.
BEACH
1
Hermosa
Beach
to.
ARTESIA
••••••••••M"
COMPTC
BLVD.
CD I
^«
1^
A"/«^ Harbor
Marina
^
ST.
•••
Southwest
Park
I90TH
Redondol
•Beach1
Figure 12. Location of real estate
developments known as Holly Glen
and Southwest Park.
Sic/*
/c
c
(X
Palos Verdesi
Estates^
[^
BLVD.
CARSON
^^
^LP
-67-
V
A J
-------
Seven of the 9 real estate dealers in Holly Glen and 6 of the 17 firms, in-
cluding the 4 largest, in Southwest Park were contacted. Most of the real estate
dealers were able to offer estimates of what homes sold for in their area of opera-
tions and by how much homes were annually increasing in value. In Southwest
Park, the average figure quoted agreed closely with the average calculated later
in the study. However, in Holly Glen, the realtors consistently quoted a figure
roughly $1,000 below the average determined later. Some real estate dealers
were able to recall instances where specific homes were sold more than once
during the 1965-1970 period of interest. The extent of such data depended en-
tirely on memory, as the brokers did not keep records in such a way as to per-
mit compilation of this type of information.
Multiple Listing Data. A more productive way of collecting property
value data was to obtain multiple listing sales often tabulated by the local board
of realtors . Several problems were encountered in using this source . First of
all, the extent of coverage of homes sold varied, depending on how aggressively
the board of realtors pushed the multiple listing service . For example, in Holly
Glen, coverage was fairly good during 1968-1970, but was quite poor during
1965, During the entire 1965-1970 period under investigation, the multiple list-
ing data for Southwest Park appeared to be much more complete.
Another problem found with multiple listing data was that reported sales
may not be final sales . In several instances, the same house would be sold twice
within three or four months and not necessarily for the same price. It was dis-
covered that such situations resulted primarily because the buyers could not
qualify for loans at the agreed upon prices. These particular situations did not
cause much problem, except that, occasionally, the final sales were not reported.
A final disadvantage associated with the use of multiple listing data was
that some boards did not compile and publish the sales data gathered by their
multiple listing service. This was the case in El Segundo. While the Hawthorne-
Lawndale and Lomita-Torrance boards of realtors compiled and published multi-
ple listing sales data, it was not generally available to the public. After it was
explained that Copley International Corporation wanted the data for a "commu-
nity problems" study, the Hawthorne data was made available. However, the
Lomita-Torrance board refused access to their information, explaining it was
for use by members only. Nevertheless, most of these data were later acquired
from other sources.
It should be noted that multiple listing data do have certain advantages.
First, they always list the date of sale rather than close of escrow, as is cus-
tomary in other sources. Second, selling price, street address, and listing
price are also always included. Finally, number of rooms, number of bedrooms,
square footage, lot size, number of bathrooms, and garage size are sometimes
provided.
-68-
-------
SREA Data. Another source of sales data for single family homes in
the test and control areas was the Society of Real Estate Appraisers. These
data have been published for Los Angeles and other cities in the United States
since 1967. As the name indicates, the purpose of the data is to assist real
estate appraisers. It is a subscription service offered only to those firms en-
gaged in regularly making first mortgage loans on single family homes . As a
part of the subscription price, subscribers are required to provide sales and
property description information on each loan approved. Sales price is exact,
but close of escrow rather than sale date is given.
The primary advantages of this source of information over the multiple
listing data were that it provided a complete description of the property sold,
the date the home Was built, and, generally, a more complete coverage of total
sales in an area. As with the multiple listing data, sales involving assumptions
of existing loans include the sales prices of the properties involved.
Comps Data. Starting in mid-1967, Comps, Inc., began tabulating
home sales data from public records on a monthly basis. Sales price can be
estimated by tabulating the transfer tax paid. Since the source of these data is
the County Recorder's Office, almost 100 percent of the sales are listed. Com-
pleteness is this source's greatest virtue.
This publication also has several shortcomings. Again, the distribution
is limited to property appraisers and lenders . However, the data are usually
sold to persons engaged in legitimate research. No details about a property,
such as square footage, number of rooms, existence of a swimming pool, etc.,
are provided.
Using the title transfer data to obtain usable sales information involves
four major problems. First, price can be determined only to the nearest $500.
Second, any transfer involving the assumption of an existing loan requires a
transfer tax only upon the balance of the loan, not on the total price of the home.
Third, as with SREA data, the close of escrow date is the only date provided.
Finally, approximately 15 percent of all transfers financed by either conven-
tional loans or assumptions of existing loans represent sales to investors with
only the investor's address being listed. This means if an investor living in
Holly Glen buys a single family home either in or outside of Holly Glen and uses
conventional financing, he may list his Holly Glen residence address as the point
of contact on the deed.5 This is not as serious a problem as it first appears,
however. Since Comps publishes this property sales data every six months, it
is likely that the investor's address will appear several times, even if he is only
moderately active. Thus, he can be identified.
-69-
-------
Evaluation of Sources
To meaningfully compare the coverage of the last three sources listed
above, the test and control areas had to be analyzed separately. Because multi-
ple listing, SREA, and Comps data were not all three available for a complete
year until 1968, the evaluation of sources was done only for 1968 through 1970.
Since many listings from each source were found not to be actual sales or the
sales price could not be determined for reasons explained above, coverage per-
centages have been listed on two bases: (1) Total Listings, i.e., all home sales
listed by each source, and (2) Usable Listings, i.e., only those transfers that
actually took place and for which the sales price could be determined. In a final
column for each basis, all three sources were combined, but the same sales re-
ported in more than one source were counted only once.
It can be seen in Table 10 that the multiple listing percentage of usable
listings in Holly Glen increased in both 1969 and 1970. Indeed, in 1970 its per-
Table 10- Single family home sale coverage in Holly Glen (1968-1970).
Multiple All Three
Year Listing SREA Comps Combined*
1968 Total Listings: #15 27 77 84
% 18 32 92 100
Usable Listings: # 12 21 36 43
% 28 49 84 100
1969 Total Listings: # 21 33 61 73
% 29 45 84 100
Usable Listings: # 14 26 40 42
% 33 62 95 100
1970 Total Listings: # 18 15 40 53
% 34 28 75 100
Usable Listings: #9 9 15 19
% 47 47 79 100
Three Year Totals (#) and Averages
Total Listings: # 54 75 178 210
% 26 36 85 100
Usable Listings: # 35 56 91 104
% 34 54 88 100
Note: No deletion rule applied within any source.
* Same sales reported in more than one source counted only once.
Source: Lawndale-Hawthorne Board of Realtors; Comps, Inc.; SREA Market
Data Center; Copley International Corporation.
-70-
-------
centage matched that of SREA. This largely reflects the Lawndale-Hawthorne
Board of Realtors' success in expanding the multiple listing service. SREA and
Comps show some variation but exhibit no definite trend.
Usable listings' percentages in Southwest Park, Table 11, show no trend
for any of the three sources. Of note here is that the SREA and the multiple
listing sources offered roughly the same coverage.
Table 11. Single family
Year
1968
1969
1970
Total Listings:
Usable Listings:
Total Listings:
Usable Listings:
Total Listings:
Usable Listings:
home sale coverage in Southwest Park (1968-1970).
#
%
#
%
#
%
#
%
#
%
#
%
Multiple
Listing
30
43
26
53
40
58
33
71
16
33
11
41
SREA
36
51
32
65
27
39
22
48
20
42
13
48
Comps
63
90
44
90
60
87
35
76
40
83
26
96
All Three
Combined**
79
100
49
100
69
100
46
100
48
100
27
100
Three Year Totals (#) and Averages (%)
Total Listings: # 86 83 163 187
% 46 44 87 100
Usable Listings: # 70 67 105 122
% 57 55 86 100
Note: No deletion rule applied within any source.
* Data for first half of each year only. Multiple listing data not available for
second half of each year.
**Same sales reported in more than one source counted only once.
Source: Lomita-Torrance Board of Realtors; Comps, Inc .; SREA Market Data
Center; Copley International Corporation.
Comparing the three year averages at the bottom of each table reveals
two important points. For usable listings, Comps provided the same coverage
in both areas, approximately 87 percent. Similarly, SREA's coverage was con-
stant but 30 percentage points lower than that of Comps. Multiple listing cover-
age of one third for one area compared to three fifths for the other was not
-71-
-------
surprising according to real estate experts. Coverage by SREA was also re-
ported to fluctuate from county to county, varying directly with the percentage
of lending institutions subscribing and providing loan data.
Examination of the ratios of usable listings to total listings disclosed
considerable variation between sources and areas. In Holly Glen, less than 50
percent of the total listings were usable in calculating average prices. One third
of the multiple listing sales either were not completed or were sales to investors
where only the investors' addresses were listed. For the SREA sources, only
one fourth of the sales were not usable for these reasons. Almost half of the
Comps recorded sales were not usable because they were sales to investors and,
more important, because the sales involved assumptions of existing loans for
which the sales prices were not available.
A similar picture emerged in Southwest Park, except only 20 percent of
multiple listing and SREA data were not usable. Almost two thirds of the Comps
information was usable. Hence, for the area as a whole, 65 percent of all obser-
vations were usable for average price calculations.
A check on the accuracy of sales price and date of sale for each source
was possible because of a large number of sales which were reported by more
than one source. Price accuracy for multiple listing data in both Holly Glen and
Southwest Park was generally good. There were only two instances out of 105
where the multiple listing price was off by more than $1,000. In all other cases,
the price differences were less than $500. In only two instances out of 123 were
SREA prices in error. However, one figure was in error by $8,000. This un-
doubtedly was a reporting error, because adding the recorded down payment to
the stated selling price yielded the actual price.
Comps data proved to be the most inaccurate, mainly due to the nature of
the recording system. Sales price is reported indirectly in terms of IRS tax
stamps affixed to the deed transfer. A tax rate table must therefore be used to
translate the dollar amount of tax stamps into sales price .6 Since $500 units
are used in determining the transfer tax due, sales price can be determined
only to the nearest $500. This is not a severe problem because most single
family homes are sold for multiples of $500 (over 70 percent, as calculated in
this study). Secondly, a home selling at the top end of a deed transfer tax bracket
pays the tax on the lower bracket. For example, a home selling for $29,000 pays
the IRS transfer tax based on the $28, 500-$29,000 bracket, not on the $29,000-
$29,500 interval. Thus, in 71 percent of the Comps sales, the true sales price
was determined by using the highest value of the IRS tax stamp implied sales
price category. In 77 percent of the cases, this method would yield a price
within $100 of the actual price. (These comments on Comps1 prices are re-
stricted to sales not involving assumptions of loans . As mentioned above, an
assumption involves a tax upon only the amount of the loan assumed and not on
the entire sales price.) The latter percentage is probably the more correct,
-72-
-------
since SREA rounds its sales figures down to the nearest $100, e.g., $28,995
would be reported as $28,900. Multiple listing data is rounded normally to the
nearest $50.
How much bias did this interpretation procedure of Comps' sales data
impart to the average price of a home? Of more than 200 sales listed by Comps
and also reported by at least one other source, the average upward bias was
$47. By deleting four selling price errors (each found to exceed $1,000), the
average bias was reduced to $38. It should be noted that, while this bias was
small for the three year period as a whole, it was much larger for any six
month period in which an error was found.?
Multiple listing source data provided the exact date a home was sold.
These dates were accurate with the only apparent chance of error being typo-
graphical .
Dates provided by SREA were generally loan approval dates . However,
since these dates were supplied by different institutions, they were sometimes
the dates of sale and sometimes the dates of recording or close of escrow.
Analysis of sales recorded by both sources showed that, in 90 percent of the
cases, the month of sale listed in the SREA report was the same month or the
following month as that recorded by multiple listing; that is, usually the loan
was approved the same month or, at worst, the month after the sale occurred.
Comps' reported date, of course, was the close of escrow or recording
date. In almost half of the cases, this date occurred by the end of the month
following the month of sale. Thus, for all houses sold during June, almost 50
percent had been recorded by the end of July. In only three percent of the cases
did the recording occur more than four months after the month of sale.
No discussion of real estate sources would be complete without mention
of the Federal Housing Administration, the Veterans Administration, and the
California Department of Veterans Affairs (Cal-Vet). Since these organizations
decide on borrowers' qualifications for the loans they finance and/or guarantee,
each possesses a wealth of information on sponsored home purchases. However,
individual sales data are generally not available due to confidentiality require-
ments . This is not to say that the data are completely beyond research use,
since, for example, Crocker was able to utilize FHA Chicago data in his study.
How accurate are the data and how good is the coverage? Sales price and
date of sale information should both be exact from all three sources. Coverage
by each source may vary tremendously. To ascertain the likely range of cover-
age in the Los Angeles metropolitan area for these latter sources, a random
sample of sales was drawn from the Comps books on southwest Los Angeles
County. For each of the 1,736 sales listed, the type of financing was noted. To
make the resulting loan distributions more informative, they were recomputed,
eliminating foreclosures and assumptions.
-73-
-------
Table 12. Distribution
FHA
VA
Conventional
Gal-Vet
Assumption of loan
Foreclosure deeds
of home loan types in Southwest
All Deed Transfers
Normal
Percent Range
4.2% 2- 6%
10.2 7-14
28.6 26-30
0.8 0- 1
54.5 50-60
1.7 1- 2
Los Angeles
County .
New Loans Only
Percent
9.5%
23.3
65.3
1.9
— —
Normal
Range
5-15%
18-30
55-75
1- 2
--
Total 100.0% 100.0%
Total no. of transfers 1,737 759
Source: Comps, Inc.
From Table 12 it is obvious that Cal-Vet would be of little use in proper-
ty value investigations. If the area of study were large enough, FHA information
might be adequate. Clearly, VA data would be the preferred source of the three
because the VA accounts for one tenth of all home financing and almost one fourth
of all new home loans. It is doubtful, however, that even VA data would have
been adequate for areas as small as those investigated in this project.
Average Price Calculations
Originally, the plan was to tabulate average, single family home prices
for each area on a quarterly basis . Due to the lack of sufficient sales data, es-
pecially in TA(Hawthorne), home sales averages had to be compiled on a semi-
annual basis. Had Comps data been available for the entire period, perhaps
quarterly averages would have been possible, although the sample sizes in
TA(H) would have been small.
Initially, sales data as far back as 1960 were sought so that the long
term trend in home value appreciation could be ascertained. It was hoped that
deviations in the trend caused by the presence of odors in Holly Glen during
1968 through 1970 would stand out. As stated above, SREA and Comps data
were not available prior to 1967. Multiple listing data for Holly Glen were not
published until 1965. Thus, the availability of data dictated that the period of
study be 1965-1970 and that averages be compiled by half years .
To tabulate a mean sales price for each half year, all three published
sources were used. This was done in order to maximize the sample size and
thereby reduce sample variability. Prior to the second half of 1967, all multi-
ple listing and SREA sales were included in the average price calculations.
For a sale to be used after June 1967, it had to have been recorded by Comps .
The purpose of this rule was to delete all those sales which, for one reason or
-74-
-------
another, either were never completed or went to real estate investors. Uncom-
pleted sales were excluded because they represented ex ante market transactions
rather than the desired actual, ex post sales. Investor sales were not wanted
because they did not represent final consumer sales and because the property
sold may have been located outside of the test and control areas. Home trans-
fers from investors to non-investor type buyers were included in determining
average sales price because such sales were for final consumption.
One criticism of the above methodology is that the data used for the per-
iod January 1965 through June 1967 were not entirely comparable to the data used
from June 1967 through December 1970. However, to make the data entirely
comparable, only the multiple listing source could have been used. It was thus
felt that any loss in comparability over time using the outlined procedure was
more than compensated by the increased number of observations in Southwest
Park and, particularly, in Holly Glen.
In order to discover which properties had been sold more than once dur-
ing the period of investigation and to facilitate matching the same sales reported
by more than one source, all home sales were recorded on index cards and then
sorted by address . It was at this stage that investor type sales became apparent.
In one case, the same home appeared to have been sold 8 times in a period of 15
months. However, reference to the buyer's name (which is provided in Comps)
showed the same person was the buyer for each sale. Based on an analysis of
the frequency of occurrence, every property listed in Comps showing two deed
transfers within four months was checked to determine if the buyer's name was
the same. In almost all cases where investors were involved, the sales data
were not included in the analysis. However, if the same house was sold several
times with the same buyer's name referenced on all but the last sale, then the
data for the last sale were included.
Where the same sale was recorded by either the multiple listing and/or
SREA, this information was transferred to the Comps card. Duplicate sale cards
were thus removed. The remaining cards were then sorted into half year periods
For reasons discussed above, date sorting was done by multiple listing dates
whenever possible. When not available, an SREA date was used if it indicated
the sale in an earlier period than did the Comps date . If only the Comps final
recording date was available, it was assumed that a July date represented a
sale during May or June and thus belonged in the average calculated for the
year's first half. Similarly, January recordings were included in the average
calculated for the second half of the previous year. From an analysis of actual
sales dates and transfer dates, this rule was judged to be conservative as
actual sales dates were probably at least 6 weeks older than the dates listed
in Comps. Any bias arising because of the one month rule used was judged as
negligible.
-75-
-------
At this point, mean home sales price was calculated for each half year
in each area. The range of sales prices for each average was also determined.
In both areas, the lower end of the price ranges stood at about 85 percent of the
mean prices. Similarly, the highest prices averaged 124 percent of the mean
home sales price in Holly Glen and 152 percent in Southwest Park.
Close examination of the range price to average price ratios drew atten-
tion to four extreme prices. Common sense indicated that one price was in
error and represented an investor purchase of property outside Southwest Park,
since it was triple the average maximum price for the control area. In addition
to this $99, 500 home sale, another was found to have sold for $69, 500. Since
this was twice as high as any other maximum in the control area, it also was
thought to be an error or an investor purchase. Both sales were deleted from
the analysis.
In Holly Glen, a $45,000 sale reported on the multiple listing sheets was
found to be $4, 500 higher than the next highest sale, while another sale at
$16, 500 was determined to be $3, 500 below the next lowest. Two realtors were
contacted who dealt in the test area; both asserted that the listed prices were in
error.
Eliminating these four sales and recomputing the range price to average
price ratios produced an average minimum price of 86 percent and an average
maximum price of 123 percent of the mean home sales price .8 Using these per-
centage relationships, the following deletion rule was developed to deal with
extreme values: In a homogeneous area, calculate the mean ratio of range price
to average price. Multiply the signed difference from one of these two resulting
ratios by 2.5 and added to one. Any home transfer having a sale price to aver-
age price ratio greater or lower than the resulting limits should be deleted. By
applying this rule to the present study, the upper and lower limits were found to
be 1.6 and 0.6, respectively.9
Such a rule is quite liberal in retaining extreme values, for it says to
retain any sale as long as it is no more than 160 percent greater than the aver-
age sales price. A similar statement holds for minimum values. By this rule,
all four suspicious values described above would be removed. Of the retained
home sales figures for both areas combined, the lowest range price to average
price ratio was 0.7; and the highest, 1.5.10 The necessity of such a deletion
rule for extreme observation is made clear by the following example. Including
the $69,500 home sale in Southwest Park for the first half of 1969, the average
price of homes in the control area would have increased by $900, even with a
sample size of 46!
The purpose of calculating average home sales prices was to estimate
market determined average prices of all homes in the test and control areas.
Since it was the pattern of growth of these averages that was important to quan-
-76-
-------
tify the economic effects of odors, extreme observations were obviously not
wanted even if they represented valid sales in the areas. Thus, the above dele-
tion rule, operating in both directions, eliminated not only incorrect and out of
area sales, but also the extreme tails of the sales price distributions.
After editing, as just described, mean sales price by half year was cal-
culated for the test and control areas. The results are displayed in Table 13
To better compare the relative growth rates of the two areas and eliminate the
$3,000 price differential, the average price data were converted to an index
basis with the second half of 1965 (II, 1965) set equal to 100. These index data
have been graphed in Figure 13. With a sample size of only 3, the value for the
first half of 1965 (I, 1965), in Holly Glen, has been omitted.
The peculiarly sharp rise and fall in average price in Holly Glen between
II, 1966 and II, 1967 was at first thought to be due to extreme values. The dis-
tribution for I, 1967 was skewed to the right, but not as severely as expected.
Using medians instead of means raised II, 1966 by $250 to $26,000 and lowered
I, 1967 by $1,000 to $27,100. The median for II, 1967 was $26,500 or less than
$200 below the mean. Each of these median values is represented in Figure 13
by an "x."
Single Source Averages. What real advantages have been achieved by
editing the three sources into one set of figures upon which an average was cal-
culated? Specifically, suppose only one source had been used with no editing;
how would the averages thus computed differ from those obtained? To answer
this question, Tables 14 and 15 were compiled for Holly Glen and Southwest Park.
In Holly Glen, the multiple listing based averages range from $1, 300
above to $1, 500 below the corresponding averages for the combined sources.
Averages for 1967 and 1968 are above whereas 1969 and 1970 figures are below
those for the combined sources. SREA averages are uniformly low. Comps
based averages range from $3,360 below to $900 above those for the combined
sources.
Contrasted to the Holly Glen multiple listing based averages, those in
Southwest Park are uniformly higher than the corresponding averages for the
combined sources. In the first half of 1970, this difference exceeds $2,000 .
Averages based on SREA data are almost uniformly above those obtained from
the combined sources, but the differences are always less than $1,000. Curi-
ously, this is the exact reverse of the situation in Holly Glen. Except in two
consecutive periods where the Comps based averages were high by $2,000, the
Comps averages very closely matched those for the combined sources.
Applying the deletion rules described above to the single source data
would undoubtedly bring the averages much closer to the figures developed from
the three sources combined. However, in dealing with very small areas such
-77-
-------
Table 13. Average sales price of homes in Holly Glen and Southwest Park for half year intervals (1965-1970).
oo
Holly Glen
Year /Period
1965
1966
1967
1968
1969
1970
I
II
I
II
I
II
I
II
I
II
I
II
No. of
Sales
3
8
8
7
14
32
25
18
22
20
10
9
Mean
Price
$32,
25,
26,
25,
28,
26,
27,
28,
29,
29,
30,
31,
730
200
480
740
240
670
680
630
380
080
690
570
Price Range
$23, 700 -$39, 000
21,500- 27,000
20,000-
25,000-
24,000-
21,000-
25,000-
23,000-
25,000-
25,900-
27,950-
28,500-
34, 500
26, 800
36, 500
32,500
39,000
37,500
38, 500
34, 300
40,500
34,500
No. of
Sales
48
38
31
5
37
46
49
38
46
36
27
25
Southwest Park
Mean
Price
$22,
22,
23,
23,
23,
24,
25,
25,
26,
26,
27,
27,
530
930
220
800
600
540
370
690
850
620
450
880
Price Range
$20, 000 -$26, 500
21,000- 27,900
21,000-
22,500-
17,000-
21,500-
22,000-
23,000-
22,000-
22,500-
21,000-
25,000-
27, 500
27,000
30,500
27,000
30,000
31,000
40,000
34,000
33,000
34,000
Source: Lawndale-Hawthorne Board of Realtors; Lomita-Torrance Board of Realtors; Comps, Inc .; SREA Market
Data Center; Copley International Corporation.
-------
128
124
120
116
112
108
104
100
96
Figure 13. Mean price of homes in Holly Glen and Southwest Park,
1965 - II * 100.
Holly
Glen
Southwest
Park
I II
1965
I II
1966
I II
1967
I II
1968
I II
1969
I II
1970
-79-
-------
Table 14. Average sales price of homes in Holly Glen by single source for half year intervals (1965-1970)
oo
o
Multiple Listing
Year /Period
1965
1966
1967
1968
1969
1970
I
II
I
II
I
II
I
II
I
II
I
II
No. of
Sales
3
8
8
6
5
5
8
6
14
7
8
10
Mean
Price
$32,730
25,200
26,480
25,700
29, 100
27,700
29, 160
29,510
29,320
28,790
29,370
31,170
SREA
No. of
Sales
--
--
--
--
8
9
15
12
15
18
12
3
Mean
Price
$ "
--
--
--
26, 680
28,110
27,010
28,390
28,290
28,600
30,880
28,900
Comps
No. of
Sales
--
--
--
--
6
33
27
16
22
21
10
8
Mean
Price
$ --
--
--
--
29,670
26,210
24,320
28,620
29,090
29,900
27,500
30,310
Actual Avg . Price
No. of
Sales
3
8
8
7
14
32
25
18
22
20
10
9
Mean
Price
$32,730
25, 200
26, 480
25,740
28,240
26,670
27,680
28, 630
29,380
29,080
30,690
31,570
Note: No deletion rules applied to data.
Source: Lawndale-Hawthorne Board of Realtors; Comps, Inc .; SREA Market Data Center; Copley International
Corporation.
-------
I
OO
Table 15. Average sales price of homes in Southwest Park by single source for half year intervals (1965-1970) .
Multiple Listing
Year /Period
1965
1966
1967
1968
1969
1970
I
II
I
II
I
II
I
II
I
II
I
II
No. of
Sales
48
38
31
5
37
7
30
13
40
15
16
1
Mean
Price
$22,530
22,930
23,220
23,800
23,820
24, 120
25,440
25,940
27,360
27,130
29,280
34,000
SREA
No. of
Sales
--
--
--
--
4
12
36
24
27
26
20
13
Mean
Price
$ "
--
__
--
23, 420
25, 120
25,760
25,710
26, 860
26,780
28,340
27,650
Comps
No. of
Sales
--
-~
--
--
12
45
44
36
37
36
26
24
Mean
Price
$ "
--
__
--
23,670
24,460
25, 520
27,740
28,720
26, 590
27,550
27,690
Actual
No. of
Sales
48
38
31
5
37
46
49
38
46
36
27
25
Avg. Price
Mean
Price
$22,530
22,930
23,220
23,800
23, 600
24, 540
25,370
25, 690
26,850
26,620
27,450
27, 880
Note: No deletion rules applied to data.
Source: Lomita-Torrance Board of Realtors; Comps, Inc.; SREA Market Data Center; Copley International
Corporation.
-------
as the ones under investigation, the resulting loss of observations could be criti-
cal. For large areas, Comps would be the best single source. (These data are
available from Comps on computer tape.) While SREA does not have the cover-
age of Comps, the data are in better form for area analysis. Comps simply lists
sales by street address, whereas SREA lists sales by map reference and then by
street address within the area referenced. (This information is also on computer
tape.)
Analysis of Average Price Data
If the presence of odors has an effect on property values, then, other
things equal, average home prices should grow at a slower rate in an odor affect-
ed community than an odor free community. This assumes not only that other
things remain equal during the period of comparison, but also that the odors are
sufficiently evident in the odor affected community and absent (or at relatively
low intensities) in the odor free (or relatively odor free) community to permit
measurement of the effect. The validity of this hypothesis was tested by com-
paring the growth rates of average sales price in Holly Glen (TA(Hawthorne))
and Southwest Park (CA(Torrance)). It was expected that average sales price in
Holly Glen was growing significantly slower than in Southwest Park.
Over the five year period from II, 1965 through II, 1970, mean sales
price in Holly Glen increased 25 percent as compared to 22 percent in Southwest
Park. These figures represent average (compound) annual growth rates of 4.6
percent and 4.0 percent, respectively — exactly the reverse of what was expect-
ed.
Since odors were thought to have become a problem in Holly Glen during
1968, growth rates were examined before and after this date . From II, 1965
to I, 1968, mean sales prices expanded at an average annual rate of 3.8 percent
in Holly Glen and 4.1 percent in the control area. During the problem period,
from I, 1968 through II, 1970, average sales prices accelerated in the test area
to 5,3 percent a year and slowed to 3.8 percent a year in Southwest Park.
To arrive at an average annual dollar growth rate, a simple regression
line was fitted to the data in both areas. In Holly Glen, mean sales prices in-
creased at an average rate of $572 per half year period or $1,144 a year. The
corresponding figures for Southwest Park were $509 and $1,018. As would be
expected, the regression lines fitted the data very closely.11
It is noted that an analysis of only multiple listing and SREA data indi-
cated reverse trends. The implication is that average sales price variations
are quite sensitive to the quality of the data.
-82-
-------
An explanation of the results is that the odor problem was too transitory
to be capitalized into property values. Complaints from Holly Glen residents be-
came severe only during 1969 and early 1970. Indeed, by late 1970, as discussed
in Chapter II, the apparent odor problem greatly diminished.
A more cogent explanation is probably that the technique itself was simply
not sensitive enough to detect the small dollar effects likely in this particular
study. The percentage growth rates were based solely on the end points. As was
seen earlier, adding one extreme figure, even to fairly large sample sizes, could
shift the average by several hundred dollars . In addition, the variances of the
observations that make up the averages are not small. This is dramatized by the
range of sale values shown next to each average listed in Table 13.
It is possible that some other factor having a different impact on the test
and control areas could have swamped any effect of odors on the growth of prop-
erty values . For example, with each community in a different city, differences
in property tax rates conceivably could have had such an impact. Investigating
this possibility, it was found that the total tax rate grew 23.0 percent between
1965 and 1970 in Southwest Park and 21.6 percent in Holly Glen. Between 1965
and 1968, the tax rate increased 3.8 percent in Holly Glen, but only 0.8 percent
in Southwest Park. In the 1968-1970 period, the tax rate rose 17.1 percent in
the test area versus 22.0 percent in the control area. Though the differences
were small, they are in the right direction to support the differential in property
value growth rates in the two areas. But, because these tax rates are so simi-
lar and have grown similarly, it is doubtful that they explain the noted different
growth rates in average sales prices. This contention was supported by the diffi-
culty encountered by the research team (and, undoubtedly, by potential buyers)
to obtain the tax rate figure at all,
It may be noted that the number of homes sold in II, 1966 and I, 1970
exhibited significant drops from earlier periods. This phenomenon was simply
a reflection of two tight money periods in which home mortgage money was ex-
tremely scarce. However, these events should have had no effect on the aver-
age price of homes sold as the impact would be uniform in both areas.
Turnover Rates. Another effect of odors in an area could be an increased
turnover rate. This stems from an hypothesis that the presence of odors may
cause sufficient annoyance to homeowners to impel them to move from the test
area.
To investigate this possibility, the number of units sold per 100 units
per period was calculated for both Holly Glen and Southwest Park. No discern-
ible pattern was uncovered, as Table 16 indicates. It could not be determined
why the turnover rates in Holly Glen were generally lower than in the control
area.
-83-
-------
Table 16. Turnover rates of homes per 100 units
1965
1966
1967
1968
1969
1970
I
II
I
II
I
II
I
II
I
II
I
II
Holly Glen
0.3
0.7
0.7
0.6
1.2
2.8
2.2
1.6
1.9
1.8
0.9
0.8
Southwest Park
2.5
2.0
1.6
0.3
2.0
2.4
2.6
2.0
2.5
1.9
1.4
1.3
Note: Holly Glen contains 1,141 homes and Southwest Park, 1, 895.
Source: Lawndale-Hawthorne Board of Realtors; Lomita-Torrance Board of
Realtors; Comps, Inc .; SREA Market Data Center; Copley International
Corporation.
A new odor problem might not cause a long term resident who has built
up strong ties in the area to move, but it might provoke a newcomer to leave.
To test this possibility, those homes sold at least twice during the 1965-1970
period were analyzed. Of those who moved into Holly Glen between 1965 and
1968, 68 percent moved out during the 1968-1970 period. In Southwest Park,
the corresponding percentage was 62. These two percentages were not signi-
ficantly different at the five percent level.
COMMERCIAL AND INDUSTRIAL EFFECTS
The opinions of retail businessmen about odors in their communities
were studied as part of the national survey of the odor problem. As a group,
the businessmen were less concerned than the residents of the same communi-
ties about odors. Only 8 out of 138 businessmen (6%) interviewed in odor
affected (test) areas felt that odors had a negative effect on their businesses.
Of these, only 2 (less than 2%) felt that the presence of odors had reduced the
annual revenue of their businesses by more than 10 percent.
-84-
-------
Since no retail firms or apartment houses were contained in either Holly
Glen or Southwest Park, no comment can be made as to the economic impact of
odors on such establishments. In general, the type of analysis used for single
family residences would be almost impossible to apply in either case. This is
due to the difficulties which would be encountered in matching apartment or retail
businesses in the test and control areas . Secondly, the success of either type of
business depends to a large degree on the competence of management. Finally,
the probable absence of large numbers of firms in both areas would deny the can-
celing effect of large numbers.
Several industrial parks are located between the oil refinery and chemical
plant complex and Holly Glen. Many firms have located in the parks small, non-
polluting manufacturing facilities and sales offices . To ascertain if any of these
facilities had suffered losses due to odors, it was planned to interview the manag-
ers of the two largest firms and at least one of the industrial parks. The manage-
ment of one of the firms would not talk with the research team because they were
considering litigation against one of the odor sources. A discussion with a repre-
sentative of the other firm proved more fruitful.
This person asserted that, on several occasions, employees had left the
plant because concentrations of hydrogen sulfide in the ambient air had drifted
into the building and the odor had made them ill. Although this situation occurred
several times, he stated that no tabulation of employees leaving because of the
odors was kept. Nevertheless, he felt that employee productivity was lowered
on days when this gas entered the plant, but no documented studies had been per-
formed .
It was found that two of the industrial parks in the area between the odor
sources and Holly Glen were managed by the same person. In response to ques-
tions by the research team, the manager stated that the older of the two parks
was developed between 1956 and 1964. Though both the oil refinery and chemical
plant were in existence long before 1956, he claimed no problem was experienced
until 1969 . One of the companies which leased facilities in the older park was
the firm considering litigation. He stated that, beginning in 1969, when concen-
trations of hydrogen sulfide and sulfur dioxide became high, the company exper-
ienced damage to their equipment and products. And, like the other firm,
employees had experienced sickness resulting in some absenteeism.
The manager implied that odors may have caused difficulty in financing,
constructing, and leasing facilities in the newer industrial park. For example,
representatives of various lending institutions have commented on the unpleas-
ant odors and taste experienced in the area. They have mentioned that effects
of contaminants in the air can be felt on the face and seen on construction steel
being used in the park. These conditions may have influenced the decisions of
lenders against the park developers with regard to the availability or terms of
development financing. In addition, the construction companies working on the
-85-
-------
park have experienced absenteeism that could be related to the presence of odors,
Finally, although potential leases have not stated that they would not locate in the
park because of the odors, the manager felt that the almost constant presence of
objectionable smells must have been considered a negative factor to some.
Unfortunately, no information has been gathered which could provide an
economic measure of the effects of odors on either of these industrial parks.
The difficulties discussed above represent possibilities based on sheer specula-
tion.
It seems likely that the lack of economic impact information would be en-
countered generally throughout the United States for two reasons:
(1) Although existing neighbors of odor causing industrial opera-
tions are offended and do incur damage, they do not choose to
publicize the fact beyond the courts of law.
(2) There has been little attention given to the collection of de-
tailed cost data that would make an evaluation of such an
impact possible.
OTHER ECONOMIC EFFECTS OF ODORS
The possibility that odors cause health effects is currently being investi-
gated by other research organizations . ^ If relationships are found, then odors
could have economic consequences in the form of medical, hospital, and other
related costs. Unfortunately, no evidence of such consequences could be located.
Odors could adversely affect tourism. However, there was no tourism
to speak of in the test area. Public beach areas were located immediately west
of the oil refinery. Since the prevailing breezes were in a westerly direction,
this area was rarely bothered by odors. Once again, no records were kept on
beach attendance which might provide a base for assessing an economic impact.
The tax base of the area could have been lowered by the presence of odors.
But, since the available methods were not sensitive enough to detect any fall in
single family property values, no effect on the tax base from residential proper-
ties could be calculated. Slower development of the industrial parks may have
caused the tax base to grow more slowly, but no dollar figure could be attached
because of inadequate records. In short, economic effects seemed possible,
but they simply could not be quantified within the scope of this study.
-86-
-------
SUMMARY AND CONCLUSIONS
If odors have an economic effect on homes, it should be capitalized nega-
tively into the price of the property. The major approach used in this project to
quantify such effects was to compare the growth rate of average property values
in closely matched test and control areas. Four major sources of data were
located:
(1) Real estate offices
(2) Multiple listing publications
(3) Society of Real Estate Appraisers
(4) Comps, Inc.
Although real estate offices kept sales information for various periods
of time, it was generally not made available to the research team. Multiple
listing information offered exact sales price and date, but lacked good coverage.
SREA data offered better coverage, but loan approval rather than sales date
was provided. Comps information, based on deed transfer records, provided
100 percent coverage, but price data was missing for roughly 40 percent of the
transfers included. Comps provided transfer dates but not sales dates . The
difference in time between the two dates averaged one to two months. Prices
could be determined from Comps only in increments of $500.
Average home sales price was tabulated by half year from 1965 through
1970. All of these sources were combined to obtain the most accurate and com-
plete data. Extreme sales data were eliminated, as were investor type sales.
Comps and SREA data were available beginning with the second half of 1967 .
Hence, the sales price averages calculated for II, 1967 through II, 1970 were
based on larger sample sizes than were the averages for I, 1965 through I, 1967.
The combined sources averages calculated for I, 1968 through II, 1970 were
compared to those based on the single sources. The Comps data were found to
be superior if the editing rules were used.
The average price of homes grew faster in the test area than the control
area for both the entire period and the 1968-1970 period when an odor problem
was in existence . Three explanations were offered for this finding:
(1) The problem was too transitory to be capitalized into prop-
erty values.
(2) The technique adopted was simply not sensitive enough to
detect the small dollar effects likely in this particular
study.
(3) Some other factor having an opposite effect than odors may
have been present.
-87-
-------
Since strong odors were present frequently in the test area beginning in 1969
(this is substantiated in part by the results of the public attitude surveys con-
ducted in TA(Hawthorne)) and since no other factors were found to offset the
effects of odors, the most plausible explanation is the insensitivity of the tech-
nique .
This study cannot be construed as proving there is no economic effect on
home values from odor problems. Although the results of the public attitude
surveys (Chapter IV) indicated that odors had not reduced the value of single
family residences, subjective evidence was encountered that odors had caused
economic losses for at least three industrial concerns. While these losses
could not be quantified, they were apparent in terms of increased absenteeism
and lower productivity of employees . No other economic effects were found.
NOTES
•'•Robert J. Anderson, Jr., and Thomas D. Crocker, "Air Pollution and
Housing: Some Findings." Paper prepared for the Institute for Research in the
Behavioral, Economic, and Management Sciences, Herman C. Krannert Gradu-
ate School of Industrial Administration, Purdue University, Lafayette, Indiana
(January, 1970), p. 2.
2Ibid., p. 3.
3In the planning stages of this study, Crocker's FHA Chicago study was
not known to CIC .
Second bath had stall shower instead of tub.
5Apartments and commercial property transfers are published separately.
6 The tax rate is 55^ per $500 of value.
7It ranged from $22 to $164 for the six month periods affected.
o
°For Holly Glen and Southwest Park, the minimum ratios were 85 per-
cent and 86 percent, respectively. Similarly, the maximums were 123 percent
and 122 percent.
90.856 - 1.000 = -0.144, -0.144x2.5=0.360, 1.000 +(-0.360) =
0.640; 1.226 - 1.000=0.226, 0.226x2.5 = 0.565, 1.000+0.565 = 1.565.
-88-
-------
10 The highest ratio reflects a sales price 140 percent above the average
sales price; then it drops to 132 percent.
11 7 o
For Holly Glen, KT = 0.89 and for Southwest Park, R^ = 0.97.
12 Technically, this is of no significance because any difference in the
level between two areas should already be capitalized in the home prices.
Interest here is thus limited to varying growth rates.
13Particularly by the State of California, Department of Public Health,
Environmental Epidemiology Unit.
REFERENCES
Anderson, Robert J., Jr., and Crocker, Thomas D. "Air Pollution and Housing:
Some Findings." Paper prepared for the Institute for Research in the
Behavioral, Economic, and Management Sciences, Herman C. Krannart
Graduate School of Industrial Administration, Purdue University,
Lafayette, Indiana, January, 1970.
Crocker, Thomas D. "Urban Air Pollution Damage Functions: Theory and Mea-
surement ." Report prepared for the National Air Pollution Control
Administration under Contract No. CPA 22-69-52 with the Regents of the
University of Wisconsin, 1970.
Ridker, Ronald G., and Henning, John A. "The Determinants of Residential
Property Values With Special Reference to Air Pollution, " Review of
Economics and Statistics, XLIX (May, 1967), 246-257.
Strodbeck, FredL., Bezdek, William E., Haller, Christian J., and Heberlein,
Thomas A. "Professional Appraisers' Judgment of the Effect of Air
Pollution on Property Values." Report prepared for the U. S. Public
Health Service under Contract No. PH 86-67-44-Neg. 5 with the Univer-
sity of Chicago, January, 1969.
-89-
-------
CHAPTER VII
DEVELOPMENT OF ASSESSMENT PROCEDURES
This chapter focuses on the basis and development of procedures to assess
the social and economic impact of odor problems in any community in the United
States. The basis for what follows evolved from the findings of the national survey
of the odor problem, the present study, and other studies dealing with community
odor problems . The development took consideration of odor complaints; the areal
extent, frequency, duration, intensity, and temporal proximity of odor perception;
the influence of meteorological conditions; and the results of comprehensive field
evaluations and public attitude surveys. Three other factors — the needs of local
authorities, the ease with which the procedures could be applied, and the costs of
undertaking the tasks involved — influenced thought in the development stage.
The initial step was to understand the situations leading to community odor
problems. Discussions, particularly with local authorities, revealed two possible
situations. They are described here as acute episodes and chronic situations.
The following definitions provide a framework for convenient reference rather
than a succinct difference in the existence of such situations. 1
An acute episode is defined as a causation of odors beyond the property
limits of a source, occurring no more than once and lasting no more than one day
in any three month period. Typically, they arise from accidents or unexpected
malfunctions involving industrial processes. Quite often, the odors are of suffi-
cient intensity to motivate many persons to complain, but of insufficient duration
to cause more than temporary discomfort or annoyance.
A chronic situation is defined as a causation of odors beyond the property
limits of a source, occurring more than once or lasting more than one day in any
three month period. Typically, they develop from industrial malpractices, includ-
ing poor housekeeping, use of faulty equipment, and inadequate process controls
in combination with supporting meteorological conditions. Persons tend to com-
plain only when the odors suddenly increase in intensity from levels normally
experienced.
This chapter is divided into four sections. The first section deals with
the use of odor complaints as a basis for the identification of community odor
problems. The second and third sections discuss, respectively, the use of
public attitude surveys as the proper basis and sensory techniques as a possible
alternative. The final section of this chapter covers possible methods for prob-
lem assessment.
-91-
-------
ODOR COMPLAINTS AS A BASIS FOR PROBLEM IDENTIFICATION
Odor complaints, like most community appeals, are almost always initi-
ated by persons who purport to be somehow aggrieved. Solicitation of complaints
by local authorities is permitted (but seldom attempted) in some areas and strict-
ly avoided in others, depending upon local policy or interpretation of state law.
The vast majority of all grievances are received by telephone and handled by in-
spectors or clerical personnel. All of the air pollution control agencies visited
during the national survey of the odor problem and the present study maintained
records of odor complaints on some type of standardized form, examples of
which are presented in Figure 14.
When, for example, a complaint is received by the County of Los Angeles,
Air Pollution Control District, a summary of the information given by the com-
plainant is recorded on a Radiophone Message Log sheet (Figure 14a) and then dis-
patched to an inspector in the field. If the complaint involves odors, code 963 is
indicated. The inspector is required to drive to the home of the complainant to
verify the presence and, if possible, the source of the odors . A Nuisance Com-
plaint Form (Figure 14b) is completed by the inspector at that time, in accordance
with the responses of the complainant to the questions on the form.
Use of Complaint Information
Traditionally, odor complaint information has been used for three pur-
poses:
(1) To alert local authorities that odors were detected at speci-
fic times and locations.
(2) To determine if local law governing the causation of odors
has been violated.
(3) To describe the conditions under which odors were detected
and to enumerate some of the effects of odors experienced
by the persons who complained.
This is, of course, to support possible court action against identified offenders.
It is in the first use that odor complaints are helpful. Clearly, they re-
duce the amount of surveillance that local authorities might need to maintain
over possible odor sources. But, contradictory to traditional practices, they
cannot be cited as conclusive evidence or used in the assessment of effects.
Consequently, they cannot be applied effectively under the tenets of the appro-
priate law — public nuisance law.
-92-
-------
Figure 14a.
TIME RECEIVED
REC'O. FROM:
830 | |t3l
DO MOT CONTACT
961 -SMOKE
962-OPEH FIRE
9S3-ODORS
96H-DUST
967-tREAKDOM
968-REPORT TO COURT
969-REPORT TO H.BO.
980-CAB SERVICE
TEN 19-COME TO HOORS.
TEN 21-TELE.HOORS.
TEN SI-TELE. NUMBER?
TEN 91 -ROLL C»LL CHS.
CODE:
Air Pollution Central District-County of UM AnplM Tl
Enforcmmt Division
RADIOPHONE MESSAGE L08
ASSIGNED TO: OH.
SOURCE:
ME DISPATCHED
ADDRESS:
COMPLAINANT:
ADDRESS:
ADDITIONAL INFORMATION:
TEN 98 - COMPLAINT NO:
SOURCE:
ADDRESS:
DISPOSITION:
? 1
§ ^
i
s
Fieiire 14b.
AIR POLLUTION CONTROL DISTRICT - COUNTY OF LOS ANGELES 40D261
434 South San Pedro Street, Los Angeles, California 90013 R. 2-69
NUISANCE COMPLAINT FORM
Date
Source com-) Name
1. plained of: )
Address
2. Complainant: Name
3. Home address
4. Business address
(Print or Type)
(Print or Type)
(Print or Type)
Phone
No.
How long
there
How long
there
(Print or Type)
5. Approximate distance from source
6. Wind direction when nuisance occurred, circle. N E S W
7, Dates and hours nuisance occurred
3. Nature of nuisance complained of, circle. Smoke, charred paper, dust, soot,
grime, carbon, add, gas, odors, (other )
9. Have you observed the above-described emissions coming from the source
complained of?
10. Have you or any member of your household become 111 from the nulstnce?
If "yes," who?
11. Describe nature of illness
12. How is your home address zoned
(Residential, Commercial or Manufacturing)
13. Has the property at this address depreciated In value due to the nuisance?
14. State any damage done to your property, furniture, automobile, clothing, etc.
IS. Will you appear in court to testify? Yes ( ) No ( )
16. Signature of complainant
(Married woman should use her own first name)
17. Complainant's remarks
INSPECTOR
BADGE
NO.
NOTICE
NO.
-93-
-------
The difficulty with using odor complaints, other than as indicators of
where odor problems may exist, lies in the probability that the complainants
do not represent the normal person in the community. Consumer and social
research studies have found repeatedly that the likes and dislikes of persons
who volunteer their opinions are different from the likes and dislikes of their
neighbors who must be solicited for their opinions.
Of similar importance, studies have found that persons who volunteer
their opinions tend to exaggerate. Thus, persons who volunteer complaints
about a particular situation may be expected to overstate their concern. An
example of this recently occurred in Sweden. Twelve hundred persons signed a
petition demanding that authorities eliminate the nuisance of offensive flue gases
caused by a sulfate cellulose factory. In response to a later survey sponsored
by local public health authorities, almost half of these persons stated they had
not experienced annoyance during the period specified in the petition. The con-
tradiction of statements could not be explained by diminished exposure since the
time the petition was submitted. The petition was rejected as a reliable mea-
sure of the extent of concern.
Need for an Effective Standard
In the application of public nuisance law, substantial and unreasonable
interference to the normal person in the community must be found. When the
presence of odors affects property values, as may happen in extreme chronic
situations, such interference is seldom in doubt. But, when odors involve mere
personal discomfort or annoyance, substantial and unreasonable interference is
difficult to find.2
What is needed is a standard against which community odor problems
can be properly identified and assessed. For the reasons just presented, such
a standard must consider the attitudes of a cross-section of members of odor
affected communities. From a recent survey, it is estimated that, at present,
only 5 percent of all state and local air quality agencies utilize such considera-
tions in their attacks on odor problems .3
The development and evaluation of a model odor control standard is the
objective of a third phase of research to be conducted by Copley International
Corporation for the Environmental Protection Agency during 1972. Since the
tentative content of such a standard was evident from the results of the first and
second phases of the work, a set of procedures was developed for use with the
ordinance and is provided in a manual attached to this report. It is felt that the
instructions are not applicable to acute episodes (defined above), but rather to
the more difficult to control chronic situations. Acute episodes involving emis-
sions of measurable concentrations of odorous substances into the atmosphere
can be dealt with as possible infractions of air quality standards governing such
substances, not as odor problems.
-94-
-------
For a fuller understanding of the subsequent material in this chapter, the
reader may wish to review "Procedures for the Identification and Assessment of
Community Odor Problems." The following sections of this chapter critique the
contents of the manual and discuss possible extensions .
PROPER BASIS FOR PROBLEM IDENTIFICATION
The identification of odor problems is requisite to determining if state or
local law has been violated. Indeed, identification should constitute violation.
The basis for decision rests not solely with the feelings of the population in an
odor affected (test) area, but in a comparison of their attitudes with the attitudes
of a population in an odor free (control) area. Such a comparison serves two
purposes:
(1) Under the assumption that both populations have equal right
to odor free air, it provides an equitable basis for decision.
(2) Since what the population in the test area says about odors
may not reflect what it actually feels about odors, it permits
an estimation of true attitudes .
Thus, the procedure for problem identification involves the use of public attitude
surveys and a means for comparing the results.
The most difficult task in conducting public attitude surveys is the delin-
eation of test and control areas . The procedure specifies the necessary steps .
Efficiency in undertaking this task evolves with experience . Assistance could be
obtained from property appraisers on the staffs of governmental agencies or those
in private practice.
The recommended public attitude survey was designed to obtain decisive
results as rapidly as possible. The instructions were written in such a way as
to clarify the necessary tasks as much as possible. For example, since it was
expected that clerical personnel would conduct the interviews, the interviewer
instructions are especially detailed.
The recommended sample sizes were predicated on an expected minimum
interview completion rate of 33 percent. They are large enough to permit a re-
presentative cross-section of a population to be analyzed without the need to
employ small sample statistics.
The questionnaire was intended to be as brief as possible. Thirteen
questions were developed from the present study to achieve the following pur-
poses:
-95-
-------
Question 1. Screening question related to the effective period of a street
address (reverse order) telephone directory.
Question 2. Introductory question aimed at whether there is concern
about various types of pollution in the respondent's neighborhood. This is to
establish the general context of the questionnaire in the respondent's mind.
Question 3. Same as question 2.
Question 4. Introduction to questions directed at the respondent's atti-
tude toward odors in his neighborhood.
Question 5. Question leading to the determination of an odor problem
index number for the area being studied.
Question 6. Same as question 5.
Question 7 . Same as question 5.
Question 8. Same as question 5.
Question 9. Question leading to a decision as to whether an odor prob-
lem exists in the area being studied.
Question 10. Same as question 5.
Question 11. Question to assist state or local authorities in identifying
source(s) of odors affecting the area being studied.
Question 12. Screening question to indicate possible bias in responses
due to respondent's association with the source(s) of odors.
Question 13. Same as question 12.
Each of the questions was carefully worded and tested to ensure understanding by
both interviewers and respondents. Testing indicated that all of the questions
could be administered to a respondent by an experienced interviewer in a maxi-
mum of three minutes.
Unlike most complaint forms used by air quality agencies (cf. Figure 14b),
the questionnaire does not solicit the exact time odor was last detected by the
residents of the test area or the direction of the wind at that time. These data
may assist local authorities in verifying the source and still can be collected
when complaints are received. But, since the questionnaire is meant to test
chronic situations which, by definition, exist over time, it would not be appro-
priate for the questionnaire to be used to solicit such data. Instead, recognition
-96-
-------
of the source is solicited directly. When odors are experienced in a community
frequently or over long duration, it is common for the residents to know or at
least suspect the source.
Use of Survey Results
Upon completion of public attitude surveys in matched test and control
areas, all questions may be tabulated. But, for odor problem identification,
tabulation should begin with question 9, "Would you say these odors have both-
ered you?"
Odor problem identification depends on the outcome of applying "z" statis-
tics . A 95 percent level of confidence (z ^_ 1.65) that more test area respondents
than control area respondents are bothered by odors is required before an odor
problem can be said to exist. Such confidence seemed optimal from the applica-
tion of "z" statistics to the results of the present study (Table 17). It was found,
for example, that odor problems existed in TB(Torrance) in December 1970
(z = 5.33) and March 1971 (z = 2.96) and TA(Hawthorne) in June 1971 (z = 5.39).4
Consistent with these findings, public concern in these communities about recent
episodes of strong oil refinery odors was high in these respective months.
If more than one odor problem is identified within the jurisdiction of the
concerned state or local agency, abatement priorities can be established by cal-
culating an index number for each problem. An odor problem index number
increases with the frequency, duration, intensity, or temporal proximity of odors
and the amount of annoyance they cause. The information flow leading to both
odor problem identification and index determination is diagramed in Figure 15.
Another use of survey results is that of estimating true attitudes. The
method for doing so is discussed in Chapter III. The survey results indicated
that in December, 81 percent of the respondents in TB(Torrance) and 37 percent
of the respondents in CB(Torrance) claimed to have noticed odors in their neigh-
borhoods in the last three months and been bothered by them . By applying the
correction formula to these results, it was estimated that only 70 percent of the
respondents in TB(Torrance) actually noticed and were bothered by odors. Quite
similarly, in June, 75 percent of the respondents in TA(Hawthorne) and 31 percent
of the respondents in CA(Torrance) made such claims. By applying the correction
formula, it was concluded that only 64 percent of the respondents in TA(Hawthorne)
actually noticed and were bothered by odors .
These estimates of true attitudes in the two identified problem areas are
not dramatic revelations of hidden truths. The estimates are not much different
from the uncorrected results . However, a dramatic impact can be seen by ex-
amining the results of non-problem areas. For example, in June, 32 percent of
the respondents in TA(E1 Segundo) and 27 percent of the respondents in
CA(South Torrance) claimed to have noticed odors in their neighborhoods in the
-97-
-------
Table 17 . Application of "z" statistics to the results of the public attitude surveys
00
I
Test Area/Control Area
Compared
December 1970
TA(Hawthorne)/CA(Torrance)
TA(X Hawthorne)/CA(X Torrance)
TB(Torrance)/CB(Torrance)
TB(X Torrance)/CB(Torrance)
TB(Glendale)/CB(Glendale)
March 1971
TA(Hawt±iorne)/CA(Torrance)
TA(E1 Segundo)/CA(South Torrance)
TB(Torrance)/CB(Torrance)
TB(Beverly Hills)/CB(Beverly Hills)
June 1971
TA(Hawthorne)/CA(Torrance)
TA(E1 Segundo)/CA( South Torrance)
No. of
Interviews
Completed
in Test
Area
68
82
54
51
51
75
75
75
50
75
75
No. of
Interviews
Completed
in Control
Area
50
101
110
110
51
50
50
75
50
75
75
No. of
Respondents
in Test Area
Who Noticed
and Were
Bothered
44 (65%)
47 (57%)
44(81%)
26(51%)
5(10%)
46 (65%)
30(40%)
51 (68%)
2< 4%)
56 (75%)
24(32%)
No. of
Respondents
in Control
Area Who
Noticed
and Were
Bothered
27 (54%)
48 (48%)
41 (37%)
41 (37%)
11 (22%)
26(52%)
15(30%)
33 (44%)
7(14%)
23(31%)
20 (27%)
Value of
Normal
Deviate (z)
1.17
1.32
5.33*
1.64
1.03
1.14
2.96*
5.39*
0.72
"X" before name of community indicates area studied in December 1970 was replaced by alternate area in March
1971. TB(X Torrance) was eliminated from further consideration.
* Indicates odor problem in test area (z >_ 1.65).
-------
Figure 15. Method of odor problem identification and index determination using
the Problem Identification Questionnaire.
Ql-
Q2
i—1. Yes
2. No-
3. Don't know-
V
Q3
Q4
.1. Yes
2. No-
3. Don't know
OPI Determination Set
~1
Problem Identification Set*
•Q9
1. Yes
2. No-
I
3. Don't know
I
V
Qll
•Q13
* Based on test and control areas being significantly different at the 95 percent
level (one-tail test).
-99-
-------
last three months and to have been bothered by them. Use of the correction for-
mula indicated that only 7 percent of TA(E1 Segundo) respondents expressed their
true feelings.
Elimination of Surveys in Control Areas
To simplify the requirements for odor problem identification, considera-
tion was given to the elimination of surveys in control areas. It was hoped that
a percentage could be established to account for the proportion of residents in
all control areas who could be expected to say they were bothered by odors.
The grounds for pursuing this possibility was the assumption that all control
areas are sufficiently common with regard to the absence of odors and, conse-
quently, that a representative sample of residents from all such areas could be
expected to express the same attitude about odors.
Table 18 provides data collected during the national survey of the odor
problem that is pertinent to this possibility. The percentage of respondents who
were bothered by odors in the control areas varied widely from 18 percent in
Tampa to 45 percent in Portland. These extremes are significantly different at
the 1 percent level (two-tail test) and cannot be considered from the same popu-
lation .
Table 18. Percentage of respondents who were bothered by odors and socio-
economic characteristics of control areas determined during the national survey
of the odor problem. *
Location of
Control Area
Portland, OR
Kansas City, MO
Cincinnati, OH
Buffalo, NY
Philadelphia, PA
Tampa, FL
San Francisco, CA
Percentage of
Respondents
Bothered
45%
32
25
42
41
18
34
Median
Family
Income
$ 5,900
7,300
12,600
6,600
9,300
9,300
5,500
Median
Education
12.5 yrs.
11.8
13.2
11.0
12.6
12.6
11.5
Median
Age of
Respondent
59.4 yrs.
51.3
46.1
53.7
48.6
48.8
52.2
* Data developed from interviews of from 160 to 180 residents in each control
area.
A similar possibility of establishing a series of percentages, each to
represent the proportion of residents in control areas of given socioeconomic
characteristics was also rejected. Although the socioeconomic backgrounds of
the Tampa and Philadelphia control areas were virtually ideal, the percentage
of respondents who were bothered by odors was also significantly different at the
1 percent level (two-tail test).
-100-
-------
Many factors could account for these differences, the most likely of
which are:
(1) Differences in odor producing potential, i.e., differences
in the number and types of odor producing industrial faci-
lities and the number of motor vehicles in operation.
(2) Differences in atmospheric dilution capacity.
(3) Possible regional differences in attitudes about odors.
The results of the present study suggested difficulties with establishing
a series of percentages at the local level. The data presented in Table 19 indi-
cate that respondents in control areas located in the same quarter of Los Angeles
may not be of the same population. For example, avoiding seasonal variations,
the percentage of respondents who noticed and were bothered by odors in
December 1970 varied from 37 percent to 54 percent in Torrance. Again, in
March 1971, the percentage of respondents who noticed and were bothered by
odors varied from 30 percent to 52 percent in Torrance. These extremes are
significantly different at the 5 percent level (two-tail test).
Table 19. Percentage of respondents who noticed and were bothered by odors
and socioeconomic characteristics of control areas determined during the pre-
sent study. *
Control Area
Percentage of
Residents
Noticed and
Bothered
Median
Family
Income
Median
Education
Median
Age of
Respondent
December 1970
CA(Torrance)
CA(X Torrance)
CB(Torrance)
CB(Glendale)
March 1971
CA(Torrance)
CA(South Torrance)
CB(Torrance)
CB(Beverly Hills)
June 1971
CA(Torrance)
CA(South Torrance)
54%
47
37
22
52
30
44
14
30
26
$12,200
12,800
11,500
7,800
12,500
13,600
12,300
15,000+
13,600
13,900
12.8 yrs.
14.2
13.0
12.7
13.1
15.3
12.9
16.0
12.9
14.5
36.5 yrs.
35.4
41.2
53.0
37.4
41.0
41.2
53.6
42.5
40.1
*Data developed from interviews of from 50 to 110 residents in each control area,
-101-
-------
The data also indicate that it would be an almost impossible task to pre-
dict the proportion of residents in a given control area from knowledge of the
socioeconomic characteristics of that area. Note that CB(Glendale) and
CB(Beverly Hills) were the lowest and highest family income control areas in-
vestigated in the Los Angeles metropolitan area. Yet, both contained relatively
fewer respondents who noticed and were bothered by odors than any of the inter-
mediate family income control areas. The same pattern held for education, and
no consistent relationship could be established for age.
It must be recalled, however, that CA(Torrance) was established not
because it represented the most odor free community having socioeconomic
characteristics similar to TA(Hawthorne), but because it was relatively odor
free and satisfied the "other things equal" criterion of being nearby a major oil
refinery. This was necessary for the property value analysis described in
Chapter VI. Of all the Torrance control areas, only CA(South Torrance) could
be considered as odor free as possible. Appropriate rules for use in future
studies would be to establish a control area at least two miles from a major odor
source and to avoid downwind locations entirely (as was the case for CA(Torrance)
to some extent in December and March).
The best possibility of eliminating surveys in control areas was found by
combining the results from CA(South Torrance) with the results from the ques-
tionnaire test area (located two miles south of CA(South Torrance)). The combi-
nation of data produced an average of 14.8 respondents who noticed and were
bothered by odors out of an average of 56.3 interviews completed, which is
26.3 percent. This possibility, to be tested again by Copley International
Corporation during the third phase of research, would require a combination
of the results of about three public attitude surveys in communities having simi-
lar socioeconomic characteristics, which are as odor free as possible.
Reduction of Effort in Test Areas
If surveys in control areas can be eliminated by the possibility just dis-
cussed, the averages established could be used to reduce the effort required to
survey test areas. This reduction of survey effort seems possible through
sequential analysis.
The fundamental difference between sequential analysis and a convention-
al survey technique is that the number of interviews completed using sequential
analysis is dependent upon the number of answers to key questions. After each
interview is completed, the results obtained are added to those already obtained
during the survey and compared with a pair of statistics previously computed.
On the basis of this comparison, one of three decisions can be made:
-102-
-------
(1) That an odor problem is identified.
(2) That an odor problem does not exist.
(3) That the results are inconclusive and that an additional
interview must be completed.
Before interviewing is begun, two sets of critical values are computed.
One set of these values indicates, for each sample size, the minimum number
of respondents who must claim to have noticed and been bothered by odors before
it can be decided that the community being surveyed has an odor problem. Such
values are, therefore, "odor problem decision numbers ." The other set of values
indicates, for each sample size, the maximum number of respondents who can
claim to have noticed and been bothered by odors before it can be decided that
the community surveyed does not have an odor problem. These values are "no
odor problem decision numbers."
In the southern quarter of the Los Angeles metropolitan area, it was
found that 26.3 percent of the respondents in a combination of control areas
said they noticed and were bothered by odors. At least 41.0 percent of respon-
dents in a test area would have to claim such an effect before an odor problem
could be identified in that area, i.e., before z = 1.65. Since the control areas
were as odor free as possible, the decimal equivalent of 26.3 percent may equi-
tably be called the "acceptable odor level (AOL)." The decimal equivalent of
41.0 percent may be called the "odor problem level (OPL)." Given an AOL and
OPL, a sequential sampling plan could be designed that is as decisive, but gen-
erally less costly than the conventional survey technique recommended in the
attached procedure manual. Such a plan would be based on the relationships:
c = -h1 + sN
c = h0 + sN
t, 7 l-a/i OPL (1-AOL)
where: hx = log -^-/log AOL(1.opL)
, /i (l-AOL)
= log
. l-AOL nnn QPL (1-AOL)
S = 10 /10g AOL (1 -OPL)
-103-
-------
and where: c = an odor problem decision number (when considered
in relation to h$ or a no odor problem decision num-
ber (when considered in relation to b^). In general,
c represents the number of respondents in the com-
munity being surveyed who say they have noticed
and been bothered by odors
a = the probability that an odor problem decision will
be reached when no odor problem exists
g = the probability that a no odor problem decision
will be reached when an odor problem exists
AOL = an acceptable odor level
OPL = an odor problem level
N = sample size
Sequential analysis was tried on the questionnaires completed in the June
public attitude surveys of TA(Hawthorne) and CA(Torrance). This was possible
several weeks after the surveys had been conducted, since the dates and times of
the interviews were marked on the questionnaires. By employing the above rela-
tionships, the sequential sampling plan given in Table 20 was prepared. By refer-
ring to the plan, it was decided that an odor problem existed in TA(Hawthorne)
after reviewing only seven questionnaires in chronological order (Figure 16a). It
was further decided that, as had been anticipated, no odor problem existed in
CA(Torrance) after reviewing only 41 questionnaires in the order in which they
were completed (Figure 16b). This would have permitted a reduction of 68 inter-
views in TA(H) had sequential analysis been employed during the survey. Such
a reduction would have been possible because of the extent of the problem in
TA(H). It is estimated that such a decision would not have been reached quite
so readily or at all during surveys of the other test areas. Attempts at apply-
ing this technique will be continued in the third phase of research.
POSSIBLE ALTERNATIVE BASIS FOR PROBLEM IDENTIFICATION
It may be possible to use sensory techniques to identify community odor
problems, but if public nuisance law is to be applied to odor cases, sensory
techniques must be shown to be reliable predictors of public attitudes about
odors. It was planned to correlate the data obtained from the use of scento-
meters and odor judgment panels with the information collected during the pub-
lic attitude surveys to establish dose/response relationships. Unfortunately,
too little was discovered, particularly in terms of day-to-day differences, to
be conclusive.
-104-
-------
Table 20
exists in
&=0.05)
, Sequential sampling plan for deciding whether or not an odor problem
a community being surveyed. (AOL = 0.263, OPL = 0.410, a = 0.05,
Sample
Size
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
No Odor
Problem
Decision No .
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
1
1
1
2
2
2
3
3
3
Odor
Problem
Decision No .
*
*
*
*
*
*
7
8
8
8
9
9
9
10
10
10
11
11
11
12
12
12
13
13
13
Sample
Size
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
No Odor
Problem
Decision No .
4
4
4
5
5
5
6
6
6
7
7
7
8
8
8
9
9
9
10
10
10
11
11
11
12
Odor
Problem
Decision No .
14
14
14
15
15
15
16
16
16
17
17
17
18
18
18
19
19
19
20
20
20
21
21
21
22
* A no odor problem decision cannot be reached until 17 interviews have been com-
pleted; an odor problem decision cannot be reached until 7 interviews have been
completed.
The principal difficulty was with the intensities of odors encountered. All
three studies took place during what may be characterized as moderately chronic
situations. Although odors were perceived on most days in Test Area A, they
were of only slight to moderate intensity. In all probability, episodes of strong
odors on the first or second day of each survey would have produced more satis-
fying results.
In terms of quarterly survey averages, the percentage of respondents in
TA(Hawthorne) who answered affirmatively to "Would you say that these odors
have bothered you?" did not vary well with the mean odor intensities recorded
by the odor judgment panel. A closer relationship was found between the per-
-105-
-------
Figure 16a. Sequential sampling chart indicating an odor problem
inTA (Hawthorne) during the June 1971 public attitude survey.
0
10
20
30
40
N
Figure 16b. Sequential sampling chart indicating no odor problem
in CA (Torrance) during the June 1971 public attitude survey.
10 -
N
-106-
-------
centage of respondents who answered affirmatively to "Have you noticed any
odors in your neighborhood in the last three months?" and the percentage of
time odor was detected by the panelists. These results suggest two possible
conclusions:
(1) The longer the time odors can be detected in a community,
the greater the number of residents who will notice them.
(2) The mean intensity of odors detected in a community is not
a good indicator of annoyance to the residents.
Unless odor intensities can be somehow related to an index of problem
perception, the use of sensory techniques may be limited to estimating the per-
centage of residents in a community who notice odors . Even so, there may be
an indirect relationship to annoyance. Table 21 shows the percentages of re-
spondents in test areas affected by oil refineries who were bothered by odors
that they noticed. Although chi-square analyses indicated that the differences
between areas were significant at the 1 percent level, differences within a par-
ticular area between quarters were not significant. This could mean that, for
a particular test area affected by a particular source, the number of residents
who are bothered by the odors caused by that source may be a constant percent-
age of those who notice the odors . Since sensory techniques probably could be
used to predict the percentage of residents who notice odors, under these very
restrictive conditions they could be used to predict the existence of an odor
problem as well. Copley International Corporation will give further considera-
tion to this possible alternative for problem identification in the next phase of
research.
POSSIBLE METHODS FOR PROBLEM ASSESSMENT
It was planned that the national survey of the odor problem and the study
of the social and economic impact of odors would provide a series of relation-
ships that could be used in the development of federal air quality, performance,
and emission standards. Yet, as this and other research progressed, it became
evident that the measurable reaction to a vast majority of odor problems was
annoyance, not other social or economic effects . Thus, a general means for
dealing with such problems must consider personal evaluations as central to
success . Still, extremely chronic situations may be encountered where, beyond
annoyance, other effects also should be measured so that totally equitable solu-
tions can be reached.
An extremely chronic situation was not encountered in any of the com -
munities investigated during the national survey or the present study. It is felt,
however, that such a situation may be assumed to exist if more than 50 percent
-107-
-------
Table 21. Percentage of respondents in test areas affected by the odorous
emissions of oil refineries who were bothered by odors that they noticed.
No. of
Test Area
No. of
Respondents
Who Noticed
Odors
Respondents
Who Were
Bothered By
Odors That
They Noticed
Percentage of
Respondents Who
Were Bothered
By Odors That
They Noticed
December 1970
TA( Hawthorne)
TB(X Hawthorne)
TB( Tor ranee)
March 1971
TA( Hawthorne)
TA(E1 Segundo)
TB(Torrance)
June 1971
TA(Hawthorne)
TA(E1 Segundo)
48
51
45
59
44
66
59
42
44
47
44
46
30
51
56
24
92%
92
98
78
68
77
95
57
of test area residents interviewed state they have been bothered "Very much" by
odors in their community.
Because of a deficiency of recorded information in most localities, deter-
mination of the types of effects to be measured is best approached with an extended
public attitude survey. A questionnaire which would fill this need appears in
Appendix E . The form is an extension of the problem identification questionnaire.
Questions 1 through 10 and 22 through 24 reproduce the questionnaire entirely.
Thus, the effects determination questionnaire could serve two purposes, viz.,
to broaden the data base for odor problem identification and to determine how
residents feel odors have affected themselves, their families, and their prop-
erties .
Methods of measuring effects are available. Social effects are measured
to some extent by the questionnaire itself. Forced changes in family living pat-
terns may be found from responses to questions 11 through 14. Although revela-
tion of health effects is solicited in questions 15 through 18, verification and
measurement of the extent of such effects would entail contacts with respondents'
physicians. The principal economic effect of odors can be measured especially
in terms of residential property value differentials . The possibility that such an
effect exists is solicited in questions 19 through 21. A description of the data
-108-
-------
required and analysis employed in this aspect of problem assessment is presented
in Chapter VI.
SUMMARY
The initial step in the development of assessment procedures was to un-
derstand the situations leading to community odor problems. Two possible situ-
ations — acute episodes and chronic situations — were defined, but only the
latter may lead to an odor problem.
The difficulty with using odor complaints as conclusive evidence of the
existence of odor problems or in the assessment of effects was discussed. It
lies in the probability that the complainants do not represent the normal person
in the community. Thus, complaints cannot be applied effectively under the
tenets of the appropriate law — public nuisance law.
In the application of public nuisance law, substantial and unreasonable
interference to the normal person in the community must be found. The iden-
tification of community odor problems is requisite to determining if state or
local law has been violated. The basis for decision rests not solely with the
feelings of the population of an odor affected (test) area, but in a comparison of
their attitudes with the attitudes of a population in an odor free (control) area.
Public attitude surveys were recommended and employed in the identification of
community odor problems.
To simplify the requirements for odor problem identification, consider-
ation was given to the elimination of surveys in control areas. The best possi-
bility of doing so was found by combining the results from the odor free control
areas investigated in the southern quarter of the Los Angeles metropolitan area.
The results of surveys conducted in Los Angeles test areas of similar socio-
economic background could be compared to the combined results. It is felt
that such comparisons could be made for at least one year before additional
control area surveys would have to be conducted.
The same combined results could be used to reduce the effort required
to survey test areas . This reduction of survey effort seems possible through
sequential analysis under which one of three decisions can be made:
(1) That an odor problem is identified.
(2) That an odor problem does not exist.
(3) That the results are inconclusive and that an additional
interview must be completed.
-109-
-------
Sequential analysis was tried on the questionnaires completed in the June
public attitude surveys of TA(Hawthorne) and CA(Torrance). Using this method,
it was found that an odor problem existed in TA(Hawthorne) after reviewing only
seven questionnaires in chronological order and that no odor problem existed in
CA(Torrance) after reviewing only 41 questionnaires in the same way.
It was planned to correlate the data obtained from the use of scentometers
and odor judgment panels with the information collected during the public attitude
surveys. Unfortunately, no conclusive dose/response relationships were dis-
covered .
A possible method for problem assessment was discussed. It would
involve an extension of the problem identification questionnaire recommended
in the attached procedure manual and, consequently, could be used to broaden
the data base for odor problem identification and to determine how residents
feel odors have affected themselves, their families, and their properties.
NOTES
•"•A "chronic situation" is meant to include not only prolonged or frequent
causation of odors, but also more than one acute episode in any three month per-
iod. The three month period was stipulated on the assumption that most persons
would be unable to recall details of previous episodes after such time .
2 William L. Prosser, Law of Torts (3rd ed.; St. Paul, Minnesota: West
Publishing Company, 1964), 599 and 602.
q
0 It could not be determined from the survey how the few agencies that
considered public opinion selected samples of residents to be interviewed.
4 If TA(Hawthorne) had been compared to CA(X Torrance) in December
and CA(South Torrance) in March, it would have been decided an odor problem
existed in the test area at both times (z equalled 2.20 and 3.43, respectively).
REFERENCES
Cederlof, Rune, Friberg, Lars, Jonsson, Erland, Kaij, Lennart, and Lindvall,
Thomas. "Studies of Annoyance Connected With Offensive Smell From
a Sulfate Cellulose Factory," Nordisk Hygienisk Tidskrift, XLV (1964),
39-48.
-110-
-------
Duncan, AchesonJ. Quality Control and Industrial Statistics. rev. ed.
Homewood, Illinois: Richard D. Irwin, Inc., 1959. See especially
Chapter VIII, "Acceptance Sampling by Attributes: Double and Sequen-
tial Fraction-Defective Sampling Plans ."
Ferber, Robert. Statistical Techniques in Market Research. New York:
McGraw-Hill Book Company, Inc., 1949. See especially Chapter VIII,
"Sequential Analysis: A New Tool for Commercial Research."
Jonsson, Erland. "On the Possibilities to Establish Air Quality Criteria for
Odours Through Studies of Annoyance Reactions.' Paper presented at
the Conference on Methods for Measuring and Evaluating Odorous Air
Pollutants at the Source and in the Ambient Air, Stockholm, June 1-5,
1970.
Prosser, William L. Law of Torts . Srded. St. Paul, Minnesota: West Publish-
ing Company, 1964. See especially Chapter 17, "Nuisance."
-Ill-
-------
APPENDICES
-113-
-------
APPENDIX A
FIELD PROGRAM PREPARATION DATA
Map of test and control areas studied in Los Angeles metropolitan area.
Topographical features of TA(Hawthorne) and TA(E1 Segundo).
Local climatological data sheets of Los Angeles, California, International
Airport for December 1970, March 1971, and June, 1971.
-115-
-------
LOS ANBELES
METROPOLITAN AREA
URBAN PORTION
.T: -T u u-^8?
Figure A-l. Locations of test and control
areas investigated during the study of the
social and economic impact of odors.
-------
I Contour lines are at 10 foot intervals. |
Figure A - 2. Topographical features of TA( Hawthorne).
~W
rT^^Ol T" jfo=
Source: U.S. Geological Survey, Venice and Inglewood Quadrangles, Los Angeles County, California
-------
Contour lines are at 10 foot intervals.
Figure A-3. Topographical features of TA( El Segundo) .
tf ' fr*** '< "-I I4— £-«*«• Vrv' -44, • .x . m«... ....... .1 P. . ...
oc
Source: U. S. Geological Survey, Venice Quadrangle, Los Angeles County, California.
-------
LOCAL CLIMATOLOGICAL DATA
U.S. DEPARTMENT OF COMMERCE
NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION
ENVIRONMENTAL DATA SERVICE
LOS 4NGELES, C*LIFDRNJA
INTERIUTIQN4L AIRPORT
DECEMBER 1970
^^^ Latitude 33" 56 ' N Longitude U6°24'w Elevation ground. 97 ft. Standard time used: PACIFIC «„„»„,,,,
&
1
2
3
^v
6
7
8
9
10
U
12
13
14
15
16
17
16
19
20
21
22
23
24
25
26
27
28
29
30
31
Temperature "F
e
£
s
2
62
62
62
69
77
77*
72
58
62
69
73
72
63
61
61
56
58
56
56
56
56
57
61
60
66
59
58
60
60
69
68
1958
Avg.
63.2
£
3
51
53
48
49
53
56
55
52
54
49
49
49
49
50
47
51
50
44
45
45
50
47
44
50
48
49
48
43
43*
47
52
1516
Avg.
48.9
S1
*
4
57
56
55
59
65
67*
64
55
58
59
61
61
56
56
54
55
54
50*
51
51
53
52
53
55
57
52
53
52
52
98
60
Avg.
56.1
a)
•fi =
Q!
5
• 2
-i
-4
1
7
9
6
-3
0
1
3
3
-2
-1
-3
-2
-3
.7
-6
-6
-4
-9
-4
-1
1
-4
-3
-4
-4
2
4
Dep.
-0.8
^
jfS.
<-s
6
49
49
44
47
44
44
46
33
50
44
31
37
47
41
47
30
43
41
43
44
46
43
39
39
35
37
44
43
43
45
46
^
Number of days
Maximum Temp.
> 90° t
0
< 32°
6
Minimum Temp.
< 32° 5 0°
0 0
Degree days
Base 65"
g
a
7A
' 8
7
10
6
0
0
1
10
7
6
4
4
9
9
11
10
11
15
14
14
12
13
12
10
8
13
12
13
13
7
5
274
Dep.
Total
416
Dep.
-141
I
6
7B
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Dep.
Total
555
Dep.
Weather types
on dates of
occurrence
I Fog
2 Heavy fog x
3 Thunderstorm
4 Ice pellets
6 Glaze
8 Smoke, Haze
9 Blowing snow
8
8
1
8
8
2 8
1 8
6
1 6
1 3 8
3
8
1 8
3
8
8
8
8
8
8
Snow
ice
lellets
or
jrounc
04AM
In.
9
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Number of days
Precipitation
> .01 inch 10
Snow, ice pellets
> 1.0 inch 0
Thunderstorms 3
Heavy fog X 1
Precipitation
Water
lent
10
0
.21
0
0
0
0
0
T
.45
0
0
0
.05
,11
• 0
,09
.07
2.23
.15
.09
.67
0
0
0
0
T
T
0
0
0
0
4.12
Dep.
1.73
Snow,
ice
pellets
In.
11
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Avg.
pres-
sure
104
m.s.l.
12
30.03
30.05
30.14
30.06
30.02
29.96
29.89
29.90
29.92
30.02
29.91
29.83
29.90
29.99
30.09
29.93
29.81
29.84
29.81
29.87
29.70
29.97
30.06
30.01
29.98
29.91
29.94
30.02
30.02
30.01
29.90
29.95
Wind
3-S
ll
13
25
26
?9
17
32
1?
14
78
0?
01
1?
14
?7
17
11
OR
26
22
16
(11
36
14
11
?6
15
11
22
34
ci.
2
a'S
II
14
2.3
5.6
1.7
2.1
2.2
1.8
.1
2.3
4.2
1.9
5.8
1.8
3.3
3.3
1.5
3.4
10.6
10.5
11.4
6.8
10.4
.7
3.4
6,2
4,1
,7
1.5
1.1
2.9
2.7
1.3
.1
13
O,
S ft
15
6.3
10.6
7.2
5.8
6.3
7.1
6.0
5.9
9.5
6.6
7.6
7.3
6.6
6.1
7.6
8.1
11.7
11.9
12.1
9.1
14.0
6.9
7.2
7.3
7.6
6.2
6.8
7.1
5.6
6.5
9.9
7.9
Greatest in 24 hours and dates
Precipitation 1 Snow, ice pellets
2.37] 18-19 1 0
Clear 10 Partly cloudy 9 Cloudy 12
Fastest
mile
$'£
16
13
35
14
1?
13
11
12
14
14
13
71
IB
16
17
3?
Ifi
3?
11
14
1,4
13
12
1ft
13
12
1,7
14
o
3
17
W/
W/
W/
W/
W/
W/
W/
W/
NW/
W/
W/
N/
M/
Vl/
E/
E/
S/
W/
W/
•til
E/
H/
NW/
VI
VI
VI
SW/
351 W/
Date: 21 +
I
Sunshine Sky cover
T)
£
m
1
18
Po«
ibie
& 0
10
month
%
It
20
t
7
3
9
10
0
2
4
3
1
9
10
9
8
9
8
0
8
7
10
9
3
0
9
5
171
Avg.
},S
0
•&-S
11
21
7
6
2
0
2
5
3
6
10
7
6
6
9
8
2
9
7
8
6
3
0
6
4
161
Avg.
5,2
Greatest depth on ground of snow,
ice pellets or ice and date
o
1
22
2;
5
in
U
12
n
14
15
17
Ifl
19
21
22
23
24
25
26
?7
?9
^0
3]
HOURLY PRECIPITATION (Water equivalent in inches!
S
S
3
! 4
g
6
7
8
9
10
11
12
14
15
16
17
18
19
20
21
22
23
24
25
26
27
26
29
30
311
A. M. Hour ending at
1
.01
.11
.02
.01
y
2
.14
y
.05
T
8
.27
,03
4
.02
.13
5
.20
6
T
T
7
T
.03
7
8
.01
.01
.01
T
.09
* Extreme temperatures for the month. May be the last
of more than one occurrence.
- Below zero temperature or negative departure from
normal.
+ > 70° at Alaskan stations.
+ Also on an earlier date, or dates.
X Heavy fog restricts visibility to 'A mile or less.
T In the Hourly Precipitation table and in columns
9, 10, and 11 indicates an amount too small to
The season for degree days begins with July for heating
and with January for cooling.
9
.03
.03
.02
.02
10
.17
.01
.01
.01
11
7
T
.03
T
12
T
T
.02
7
T
P. M. Hour ending at
1
T
T
7
T
.08
7
2
.01
7
.01
.01
.04
T
3
.03
.01
.01
.02
4
Subscription Price: Local Climatolog-
ical Data $ 1.00 per year including
annual summary if published. Single
copy: 10 cents for monthly summary;
15 cents for annual summary. Checks
or money orders should be made payable
and remittances and correspondence
should be sent to the Superintendent
of Documents, U . S . Goverment Print-
ing Office, Washington, D. C. 20402.
Data in columns 6, 12, 13, 14, and 15 are based on 8 l
observations per day at 3-hour intervals. T ^--.^f^, t-ha^ rYti <* iq an official
Wind directions are those from which the wind blows. I cer tlly tnat tnis is an °"^" <;
Resultant wind is the vector sum of wind directions publication of the National Oceanic
and speeds divided bv the number of observations. and Atmospheric Administration, and
Figures for directions are tens of degrees from true is compiled from records on file at
North ;i.e., 09 = East, 18 = South. 27 = West, 36 = North. the National Climatic Center, Ashe-
and 00 = Calm. When directions are in tens of degrees . , , , , ,_ jlina 28801.
in Col. 17, entries in Col. 16 are fastest observed vine, i , TO. "" /• .
1-minute speeds. If the / appears in Col. 17. speeds . •••••••' , / jl fl
«re gusts. wMf»-~H- ?^ff*~r
Any errors detected will be corrected and changes in ,.•,„•,.*/•,.'..
summary data will be annotated in the annual summary. Director, National Climatic Center
.05
.02
.02
,
5
7
.09
6
7
.07
.02
7
.36
T
8
.28
T
9
.26
10
.36
.04
11
.05
.32
12
7
T
7
.44
.01
SUMMARY BY HOURS
AVERAGES Resultant
4> M
E 1
u3 8-§
j'td >,\
ez^
01 5
04 5
07 7
10 6
13 5
16 5
19 5
22| 5
u
I1B
iJ M M
cd OJ
U kJ
t/i a
29.9
29. 9<
29.9
29.9
29.9
29.9
29.9
29.9
Temperature
•H
i 53
5C
i 50
) 57
> 62
1 59
t 56
i 54
b.
.a
S J3
48
46
46
50
52
53
51
49
&
(14
So
S
o
43
42
41
42
43
46
46
44
H
V X
CB .^
Q) §
A £.
72
74
72
61
54
64
73
70
"8
i.^ I
4 "*• S
i °
6.4 06
6 • 9 Ql
7.1 08
6.4 06
9.9 2*
10.4 26
7.9 26
6.2 05
It
3,3
4 • t
4.0
2.7
4.7
7.5
4,2
2.1
S
a
J
5
6
7
6
9
10
11
12
13
14
15
16
17
16
19
20
21
22
23
24
25
26
27
26
29
30
31
USCOMM — NOAA — ASHEVILLE 400
-------
OBSERVATIONS AT 3-HOUR INTERVALS
1
01
04
07
10
13
It
It
22
01
04
07
10
13
16
19
22
01
04
07
10
13
16
19
22
01
04
07
10
19
16
19
22
01
04
07
10
16
19
22
01
04
07
10
13
16
19
22
01
04
07
10
13
22
01
04
07
10
13
16
19
22
01
04
07
10
13
16
19
22
01
04
07
10
13
16
19
22
01
04
07
10
13
16
19
22
*
H
*"
*
i
10
9
5
7
7
9
7
3
2
9
1
0
0
0
0
1
1
e
3
2
a
7
8
1
0
0
0
0
0
0
0
1
0
g
0
7
9
10
7
10
10
10
10
10
10
10
10
9
10
9
9
7
7
9
2
5
4
2
2
1
10
10
9
7
6
9
5
3
0
4
7
0
2
5
1
1
7
8
10
9
3
0
0
15
-z
UNL
45
50
UNL
50
30
250
250
UNL
UNL
250
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
12
UNL
UNL
21
35
UNL
17
22
200
6
4
55
35
18
40
40
40
25
200
25
50
UNL
UNL
UNL
UNL
UNL
UNL
250
250
250
230
UNL
UNL
UNL
UNL
UNI
UNL
UNL
UNI
UNL
UNI
UNL
UNL
UNL
150
150
120
UNL
UNL
UNL
UNL
VISI
kILIT*
.i
15
5
12
6
10
20
It
15
8
12
li
3
5
6
5
15
15
90
10
10
15
12
10
12
10
30
a
15
25
15
12
2
2
4
4
15
10
10
4
3
1
1
1
2
15
15
10
12
»
1
10
15
15
12
4
6
i;
15
15
15
15
15
50
50
40
35
i;
15
15
15
6
3
6
12
15
15
10
12
20
5
4
6
7
5
Ss
8
8
MtkTHER
OAY 01
ui:
i A
tttATHEft
DAY 02
DAY 05
KH
KH
OAY OB
f
F
KH
KH
OAY 11
OAY 14
DAY 17
RW
DAY 20
ft*
R»
RK
OAY 23
DAV 26
OAY 29
DATA
ational
icrof ic
and cc
ding, f
NOELES
TEMPERATURE
ee
<
56
57
59
57
54
57
53
57
68
75
60
61
53
53
57
58
53
51
63
72
67
63
57
53
51
51
57
60
59
56
55
52
S3
57
55
53
53
49
46
46
51
54
53
53
48
45
44
54
60
57
52
37
56
54
46
46
43
55
59
A
53
56
52
50
48
52
48
47
53
58
57
52
53
53
55
56
43
42
51
2
6
0
4
a
51
52
49
45
50
50
48
47
45
46
44
3
7
9
45
43
41
46
49
49
45
49
49
45
43
40
46
52
srs-
2
o
Z
**
a
511 93
56
45
43
42
47
44
36
39
44
54
43
52
53
53
54
54
31
25
37
24
47
35
38
43
45
43
33
47
43
41
40
37
42
42
39
43
45
42
40
38
36
36
41
37
40
44
41
39
37
37
45
96
60
60
64
69
72
46
35
33
ei
52
87
100
100
90
87
87
4
1
3
2
eo
54
49
54
60
62
43
80
60
59
62
55
77
96
77
74
72
90
93
80
51
41
55
57
53
69
77
77
79
51
60
WIND
s
25
26
26
36
04
04
09
36
05
29
24
04
08
11
14
25
24
06
36
05
30
34
25
06
34
24
27
24
13
25
26
25
26
27
31
05
07
06
09
11
10
04
06
04
06
28
30
01
31
26
23
09
11
14
09
26
Q«
£1
**
5
10
16
5
5
*
3
7
4
7
7
4
7
6
6
4
4
3
7
6
B
B
6
6
5
9
11
4
T
e
3
11
16
16
15
7
10
5
10
B
9
5
6
3
6
4
11
10
5
9
4
5
2
5
5
5
3
5
he, or paper copies of
sts should be addressed
sheville, North Carolin
CALIFORNIA yEAR „ H(.
et
fi
f
0
6
3
2
1
1
7
10
10
9
0
4
e
5
3
0
0
0
0
1
7
10
1
10
10
10
4
2
1
9
a
10
10
10
10
10
10
10
8
10
10
10
10
7
9
10
10
7
9
9
0
0
0
4
6
10
10
rds
z-=
-i
UNL
UNI
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
35
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
250
250
UNL
UNL
UNL
UNL
UNL
UNL
30
2B
23
15
60
29
15
IS
IS
280
290
250
200
200
200
150
UNL
50
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
on i
VISI
BILITV
3d
fi
10
12
25
20
20
30
IS
15
12
15
90
50
30
12
12
1
20
25
30
15
15
10
12
90
30
15
7
4
7
4
5
7
7
5
15
15
IS
15
30
8
4
4
15
6
10
10
10
IS
12
15
30
30
20
30
15
15
7
10
IS
15
lit
||
8
8
C£
DAY 03
DAV 06
OAY 09
LKH
H
GF
OAY 12
KH
DAY 15
KH
KH
KH
OAY 18
RW
R
R
R
DAY 21
R
RW
RW
«W
DAV 24
OAY 27
DAV 30
KH
KH
KH
KH
KH
n be f i
TEMPERATURE
b.
oe
<
52
SO
49
sa
61
61
56
55
60
57
56
73
6B
59
61
58
57
57
60
61
62
59
56
58
52
50
63
72
57
50
50
57
61
55
53
49
45
45
55
53
49
4B
S3
S3
53
55
55
52
52
52
51
51
50
54
60
60
56
53
57
57
54
52
49
48
47
53
64
58
5t
ii
41
46
54
53
51
51
50
49
46
57
57
56
54
56
56
56
57
55
55
48
49
44
42
41
47
51
55
47
47
51
54
52
51
44
42
42
47
47
46
50
50
51
51
46
46
46
44
43
45
46
46
47
45
51
53
49
47
46
45
44
52
54
52
tf
44
41
47
46
47
47
40
40
39
44
54
49
55
55
55
55
51
50
35
41
25
28
28
26
27
54
44
44
46
46
50
49
38
38
38
41
45
45
47
49
48
48
45
45
44
36
35
35
35
37
36
46
49
44
41
44
41
40
41
SO
49
**
K
74
71
60
58
72
75
48
53
53
35
84
65
90
93
93
84
70
65
41
57
26
40
43
25
19
90
90
80
80
67
63
93
96
66
77
77
64
86
69
eo
83
77
77
77
77
74
57
57
49
39
49
53
67
75
69
66
86
77
77
43
75
76
WIND
S
03
05
25
26
25
07
35
07
06
26
23
07
10
09
22
26
27
27
36
03
04
12
10
05
20
27
22
11
09
07
07
14
?^
26
21
04
05
12
14
10
09
IB
27
28
29
26
26
32
01
02
01
09
35
03
25
25
32
31
11
08
09
09
26
01
Co: Director, National Climatic Center
a 28801.
NTH: 7° «
S?
»*
7
4
10
10
5
4
6
6
4
9
5
5
4
6
3
13
14
13
6
7
7
1
7
10
9
4
3
t
t
13
15
12
13
15
11
It
16
13
11
12
7
t
8
6
5
6
5
6
8
8
6
6
7
t
11
10
4
CEILING COLUMN—
UNL indicates an unlimited
ceiling.
WEATHER COLUMN—
Tornado
T Thunderstorm
Q Stiuall
R Rain
RW Rain showers
ZR Freezing rain
L Drizzle
ZL Freezing drizzle
S Snow
SP Snow pellets
1C ice crystals
SW Snow showers
SO Snow grains
IP Ice pellets
A Kail
F Fog
EF Ice fog
GF Ground fog
BD Blowing dust
BN Blowing sand
BS Blowing snow
BY Blowing spray
K Smoke
Haze
Dust
D
WIND COLUMNS—
Directions are those from
which the wind blows, indi-
cated in lens of degrees
from irue North; i. e., 09
for Easi, IS for South. 21
for West. Entry of 00 in
the direction column indi-
cates calm.
Speed is expressed in knots;
multiply by 1.15 to convert
to miles per hour.
-120-
-------
LOCAL CLIMATOLOGICAL DATA
U.S. DEPARTMENT OF COMMERCE
NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION
ENVIRONMENTAL DATA SERVICE
LOS ANGELES' CALIFORNIA
1NTEKN4.T1QNAL AIRPORT
MARCH 1971
^^^ Latitude 33° 5S' N Longitude m° „• w Elevation ground) 97 ft. Standard time used: PACIF,C HBAN *23174
S
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
i
17
18
1 19
20
21
22
23
24
25
26
27
28
29
30
31
Temperature °F
E
1
2
60
59
66
62
65
72
70
62
59
58
59
60
67
61
66
74
67
77*
67
61
60
61
60
62
60
63
62
62
65
65
64
Sum
1976
Avg.
63.7
g
i
3
46
39*
43
47
49
49
50
46
49
50
47
50
53
46
45
50
50
52
49
52
51
52
53
54
54
56
55
53
51
53
50
Sum
1544
Avg.
49,8
Average
4
53
49*
55
55
57
61
60
54
54
54
53
55
60
54
56
62
59
65*
58
57
56
57
57
58
57
60
59
58
58
59
57
Avg.
56.8
_
* i
IE
S£
5
-3
.7
-1
-1
0
4
3
-3
• 3
-3
-t
-2
3
-3
-1
5
2
8
1
0
-1
0
0
1
0
3
1
0
0
1
-1
Dep.
-0.2
+,
It
II
6
22
18
25
45
40
30
31
43
47
47
48
51
45
41
46
45
51
51
50
50
48
50
51
51
52
55
54
54
54
52
52
AvS-
45
Number of days
Maximum Temp.
590° t
0
==32'
0
Minimum Temp,
< 32° z 0°
0 0
Degree days
Base 65°
S
1
7A
12
16
10
10
8
4
5
U
U
11
12
10
5
U
9
3
6
0
7
8
9
8
8
7
8
5
6
7
7
6
8
Total
248
Dep.
-16
ToWl
1211
Dep.
-236
00
1
O
7B
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Total
0
Dep.
Total
31
Dep.
Weather types
on dates of
occurrence
2 Heavy fog X
3 Thunderstorm
4 Ice pellets
5 Hall
6 Glaze
7 Duststorm
8 Smoke, Haze
9 Blowing snow
8
8
a
8
8
8
8
8
8
1
8
8
8
2 8
8
8
8
8
8
8
8
1 8
8
2 8
1 8
8
Snow
ice
wlleis
or
Jrount
at
04AM
In.
9
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Number of days
Precipitation
3 .01 inch 1
> 1.0 inch 0
Thunderstorms 0
Heavy fop X 2
Precipitation
Water
lent
In.
10
0
0
0
0
0
0
0
D
0
0
0
T
.23
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Total
.23
Dep.
-1.56
Snow,
ice
pellets
In.
11
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Total
0
Avg.
station
pres-
sure
In
104
feet
m.s.l.
12
30
10
10
79
?9
30
79
79
10
30
79
79
29
30
30
29
29
79
79
29
30
30
29
79
29
29
79
29
29
29
29
.01
.?n
.10
.84
.97
,07
.94
.91
.07
.01
.96
.9?
.84
.09
.07
.96
.84
.97
.97
.91
• 01
.01
.94
.AD
.95
.97
.98
.39
.83
.89
Wind
Resultan
n
34
10
26
11
79
19
71
71
70
77
78
73
77
1?
1ft
76
76
76
26
76
74
77
76
75
76
24
16
26
CL.
Resultan
speed m
14
U.3
3.3
3.8
3.9
5.4
2.7
3.9
2.4
3,6
5,3
1,9
3,2
12.6
3.0
3.0
5.2
3.9
.1
7,0
6.3
4,8
7.5
4.5
4.7
2.6
11.2
7,4
4,2
6,5
2.5
6,2
h
15
14.7
6.8
8.5
8.9
10,2
7.1
7.9
7.9
9.4
7.8
6.9
9.4
16.1
9.2
6.5
7.2
7,8
8.1
9.4
6.8
7.8
8.1
5.9
7.3
6.5
11.5
8.2
6.6
7.6
7.6
6.9
Fastest
mile
If
w c
16
40
18
11
10
17
17
7?
18
in
21
17
7.5
18
17
7.0
17
22
7.3
23
17
16
15
26
71
18
18
23
20
Direction
17
N/
SW/
W/
W/
N/
W/
M/
W/
W/
W/
H/
s/
NW/
M/
W/
W/
E/
E/
W/
SW/
w/
w/
w/
w/
w/
M/
W/
SW/
H/
W/
For the month:
29
.96
Greatest in 24 :
Precipitation
.231 12-13
26
3.9
8.4
lours and dates
Snow, ice pellets
ol
40] N/
Date: 01
Sunshine
•0
£
II
18
Total
Possible
1|
sa
£•3
19
%
for
momh
• Sky cover
Sunrise t
sunset
20
0
0
5
10
5
9
2
5
0
10
0
6
7
4
5
6
10
9
10
8
9
6
5
9
0
Sum
147
Avg.
4.7
3
•a-6
II
s'i
21
0
o
9
3
2
0
7
1
6
6
7
8
10
9
10
8
9
5
4
8
1
Sum
141
Avg.
4.5
Greatest depth on ground of snow,
ice pellets or ice and date
o T
S
&
•)•>
1
5
9
10
11
12
13
14
15
•m
171
1ft
19]
71
77
71
74
75
?A
77
?B
79
10
31
Clear 11 Partly cloudy 11 Cloudy 9 1
HOURLY PRECIPITATION (Water equivalent in inches)
2
a
i
2
3
4
5
6
7
a
9
10
11
12
13
14
15
T?T
17
18
20
21
22
23
24
25
26
27
28
29
30
_3_L
1
,04
— — |
.12
3
.06
A.
4
.01
M. Ho
— 5 —
T
ar endi
6
T
ngat
7
Q
o
" Extreme temperatures for the month. May be the last
of more than one occurrence.
— Below zero temperature or negative departure from
normal.
t > 70" at Alaskan stations.
•t- Also on an earlier date, or dates.
X Heavy fog restricts visibility to >/4 mile or less.
T In the Hourly Precipitation table and in columns
9, 10, and 11 indicates an amount too small to
measure.
The season for degree days begins with July for heating
and with January for cooling.
' 9
10
12
— —
Subscription Price: Local Clitnatolog-
ical Data $ 1,00 per year including
annual summary if published. Single
copy: 10 cents for monthly summary;
15 cents for annual summary. Checks
or money orders should be made payable
and remittances and correspondence ^
should be sent to the Superintendent 1
of Documents, U. S. Goverment Print-
ing Office, Washington, D. C. 20402.
Data in columns 6, 12, 13, 14, and 15 are based on 8 '
observations per day at 3-hour intervals. , „__..-• *„ «.uof- t-hi c i c an of fi rial
Wind directions are those from which the wind blows. I P^Clfy that this is an °"icial
Resultant wind is the vector sum of wind directions publication of the National Oceanic
and speeds divided by the number of observations. and Atmospheric Administration, and
Figures for directions are tens of decrees from true is compiled from records on file at
North ; i.e., 09 = East, 18 = South, 27 = West. 36 = North. the NarJ -- -' "1-'matic Center, Ashe-
and 00 - Calm. When directions are in tens of degrees viVle — 191— olina 28801.
P.M.
lour e
ding
T
t
1 — g —
T
9
in
in
U1
12
T
SUMMARY BY HOURS
AVERAGES Resultant
II"
01 4
34 6
07 6
10 5
13 4
16 4
19 4
22 4
iJ M (-
4 55
r 54
U.
9
« y
3 J3
49
48
47
52
54
53
51
51
\°
o
45
44
42
44
47
46
47
47
1$ 1^ 1
«llM
ai 3 S O
as: *
78 5.9 06
60 5.501
76 6,8 03
62 8,1 27
62 12.2 29
64 12.7 26
77 9.2 25
78 6.7 22
TJJ=
M E
1.2
1.1
1.7
2,6
10,8
11.2
8.0
2,4
3
1
2
3
4
S
6
7
8
9
10
n
12
13
14
,15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
in Col, 17, entries in Col. 16 are fastest observed » LiiS- / — -
1-minute speeds. If the / appears in Col. 17. speeds . . / // /y
are gusts. */„£&£_ /f. 7S4*-f*««Vr/
Any errors detected will be corrected and changes in ' - ' ' ft [ nor-r>uu ««AA .CUI-UTI Tf
summary data will be annotated in the annual summary. Director, National Climatic Center UM.OMM — NUAA — AbHLVlLLt $00
-------
OBSERVATIONS AT 3-HOUR INTERVALS
K
gf
>>-
*
0
0
0
0
0
0
0
0
0
0
0
0
0
4
3
0
0
0
0
0
> 0
> 0
I 0
1 10
i 10
i 10
> 10
) 10
i 10
J i
i 0
10
t 10
T 3
) 5
> 1
i 0
» 0
! 0
0
t 0
7 7
» 10
9 10
S 10
> 10
i 10
I 6
; 10
> 5
> 1
7
10
1 10
10
10
10
10
10
10
10
10
10
10
4
6
7
6
3
1
0
2
0
0
0
0
2-3
=1
UNL
UNL
UNL
JNL
UNL
UNL
UNL
UNL
UNL
UNL
UNI
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
15
16
19
23
UNL
UNL
UNL
UNL
11
11
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
Z50
250
UNL
220
UNL
1
UNL
UNL
6
12
12
15
9
9
15
15
17
23
16
13
12
15
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
VlSI
H
15
15
90
30
20
12
12
10
15
12
10
6
8
12
10
10
15
15
50
40
40
to
15
15
7
7
5
7
7
7
7
8
6
10
12
20
35
90
30
30
10
8
IS
8
15
25
15
12
6
0
7
12
8
4
2
2
2
4
5
4
1
2
3
4
4
5
5
10
7
5
7
7
6
12
10
?r
4
B
8
e
B
V.EVTHER
DAY 01
DAY 04
KH
DAY o?
DAY 10
KH
DAY 13
R
DAY u
OAY 19
KH
F
DAY 22
KH
KH
KH
KH
KH
OAY 25
KH
KH
KH
KH
KH
KH
KH
OAY 26
KH
KH
DAY 31
KH
KH
TEMPERATURE
lb
ad
M
<
47
55
58
60
54
51
51
49
60
58
54
53
55
67
67
62
56
54
51
52
52
58
56
52
51
53
53
57
62
66
37
51
71
62
60
56
55
50
61
54
52
53
58
55
55
55
57
59
59
57
61
57
53
53
61
63
62
56
56
§1
41
43
43
42
38
36
48
46
51
53
51
50
47
50
51
51
50
49
49
48
52
51
49
49
51
51
54
50
49
55
53
53
53
46
53
54
53
53
53
54
56
56
55
57
56
51
52
56
57
58
53
52
Sir-
32
26
23
14
07
07
45
42
42
48
48
47
22
31
41
46
47
47
47
47
50
49
51
31
48
50
46
50
51
47
52
51
51
51
51
51
53
54
53
54
55
50
51
53
53
55
50
49
3
D
*f
_l
K
56
33
26
16
15
17
80
77
52
70
EO
60
18
26
46
69
77
86
67
86
90
86
80
27
90
65
60
ao
86
89
78
93
78
86
86
96
80
91
84
87
78
93
90
93
75
70
78
80
78
WIND
§
29
34
35
35
35
15
11
11
19
25
24
22
08
02
27
26
23
24
19
25
08
26
27
27
35
28
25
26
23
04
27
25
25
26
34
36
11
27
25
26
25
22
00
24
26
26
26
24
01
a«
S3
6
10
18
22
19
13
9
4
5
5
5
11
18
8
6
4
10
12
10
3
6
10
4
16
17
12
17
11
11
4
5
10
7
6
3
7
11
7
e
12
3
3
6
7
7
B
e
4
4
0
5
7
U
U
6
3
8!
>•>-
5S
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
10
0
0
0
0
3
a
1
0
0
7
9
3
0
0
10
3
0
0
0
9
6
7
10
10
10
10
0
7
9
B
3
0
A
in
a\
Ft
STA
£•=
2o
-•S
"I
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNI
UNL
UNL
UNL
UNL
UNL
UNL
UNL
220
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
290
UNL
UNL
UNL
UNL
200
UNL
UNL
UNL
UNL
a
B
10
1*
25
25
UNL
UNL
33
35
UNL
UNL
• lV?TV
1
10
10
6
12
25
15
15
7
7
30
30
40
30
15
15
10
5
10
10
10
7
7
4
7
15
15
12
8
25
15
12
10
10
7
10
8
5
5
7
1
5
4
5
20
20
20
1
7
I?
-I
B
8
WEATHER
DAY 02
KH
DA* 05
DAY 08
KH
DAY 11
KH
KH
DAY 14
DAY 17
DAY 20
KH
KH
DAY 23
KH
DAY 26
KH
KH
KH
DAY 29
KH
TEMPERATURE
h.
K
tH
<
43
39
54
51
98
51
52
If
59
61
60
55
53
57
61
59
53
52
51
58
59
58
53
47
58
60
60
53
52
55
51
66
66
60
55
54
54
53
56
57
56
63
62
58
57
60
56
DD1T1ONAL DATA
U.
si
31
31
39
43
43
46
41
46
51
51
49
49
53
56
54
51
48
53
55
54
45
52
49
49
50
53
90
60
55
53
52
52
51
53
55
54
59
58
56
56
54
es-
1
05
OS
14
15
20
22
41
24
29
41
43
43
45
49
52
50
49
46
49
51
50
43
46
37
46
51
49
55
51
51
51
50
51
53
53
56
56
55
55
53
I
=
xf
*
18
21
36
21
23
25
69
34
32
48
54
64
74
7S
72
72
86
83
72
75
75
86
65
43
77
86
93
68
72
86
90
86
83
83
87
90
78
81
90
93
90
WIHD
I
20
04
11
16
22
25
26
01
32
30
24
23
21
14
12
24
26
26
12
19
25
26
10
08
26
26
08
07
11
23
15
08
27
24
03
27
25
27
25
24
25
27
24
24
o.
W
6
3
5
2
10
12
5
4
4
13
15
13
9
7
4
6
10
10
7
6
3
12
9
3
7
B
11
13
B
6
5
10
B
6
B
3
3
2
7
8
3
15
15
15
10
10
•
SI
f
0
0
0
0
1
5
2
0
0
2
1
1
1
0
0
10
0
2
0
7
10
10
10
0
0
0
0
0
0
0
10
2
3
10
4
10
10
10
9
B
10
10
>
10
7
10
O-=
Zf
h
u c"
6 S"
* to
3
4
11
12
6
7
3
10
9
5
0
4
11
10
6
4
7
5
6
6
6
3
NOTES
CEILING COLUMN-
UNL indicates an unlimited
ceiiing.
WEATHER COLUMN-
Tornado
Thunderstorm
Squall
Rain
RW Rain showers
ZR Freezing rain
L Drizzle
ZL Freezing drizzle
S Snow
SP Snow pellets
1C Ice crysials
SW Snow showers
SG Snow grains
IP Ice pellets
Kail
Fog
Ice fog
GF Ground fog
BD Blowing dust
BN Blowing sand
BS Blowing snow
BY Blowing spray
" Smoke
Ha?e
Dusi
WIND COLUMNS—
Direct ions are those from
which (he wind blows, indi-
cated in lens of degrees
from irue North; i. e.. 09
For East, 18 for South, 27
for West. Entry of 00 in
the direction column indi-
cates calm.
Speed is expressed in knots;
multiply by 1.15 to convert
to miles per hour.
-122-
-------
LOCAL CLIMATOLOGICAL DATA
U.S. DEPARTMENT OF COMMERCE
NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION
ENVIRONMENTAL DATA SERVICE
LOS ANGELES; CALIFORNIA
INTERNATIONAL AIRPORT
JUNE 1971
^^^| Latitude 33° 56' N Longitude lle° 24' w Elevation igroundi 97 ft. Standard time used: P4CiFlc WBAN #23174
*
w
1
1
2
3
4
5
6
7
9
10
11
12
13
-it
fib
16
17
rB
19
20
21
22
23
24
25
26
27
28
29
30
Temperature °F
E
_E
cd
*
2
67
68
66
67
69
69
68
65
65
67
70
69
70
76
70
69
69
70
70
77
77*
72
73
71
72
72
71
72
72
70
2103
70.1
|
e
S
3
50*
56
55
54
57
60
58
59
59
58
59
57
59
59
60
57
60
61
58
62
61
63
60
64
64
63
63
62
61
59
1778
59.3
*
B?
&
*•
4
59*
62
61
61
63
65
63
62
62
63
65
63
65
68
65
63
65
66
64
70*
69
68
67
68
68
68
67
67
67
65
64.7
a> S
ss
t; c
§, E
Ol"
5
-5
-2
-3
-3
-1
1
-1
-2
-I
1
-1
1
4
0
-2
0
I
-1
5
3
2
1
2
2
1
0
0
0
-3
'0.1
sJ
tfS,
<1> %
<-D
6
49
51
52
54
56
56
55
56
55
55
57
57
58
60
61
60
60
59
60
61
62
61
61
61
61
60
60
57
58
58
58
Number of days
Maximum Temp.
^ 90° t
0
532°
0
Minimum Temp.
< 32° <; 0°
0 0
Degree days
Base 65°
c
CD
£
S
7A
6
3
4
4
2
0
2
3
3
2
0
2
0
0
0
2
0
0
1
0
0
0
0
0
0
0
0
0
0
0
34
-20
Total
1566
Dep.
-233
'4
7B
0
0
0
0
0
0
0
0
0
0
0
0
0
•3
0
0
0
1
0
5
4
3
2
3
3
3
2
2
2
0
33
Total
100
Dep.
Weather types
on dates of
occurrence
1 Fog
2 Heavy fog x
4 Ice pellets
5 Hall
6 Glaze
7 Duscstorm
9 Blowing snow
a
8
8
8
B
1 8
8
8
2 8
2 8
8
8
2 8
8
1 8
8
8
8
a
8
a
8
Snow
ice
pellets
or
^rounc
at
04AM
In.
9
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Number of days
S .01 inch o
> 1.0 inch o
Thunderstorms o
Heavy fog X 3
Precipitation
Water
lent
In
10
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.00
.0.05
Snow,
ice
pelleis
In.
11
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Avg.
pres-
sure
In
Elev.
m
29
29
79
79
29
29
79
29
29
79
79
29
29
29
79
29
29
79
79
79
29
79
79
79
29
29
29
29
29
29
29
S.I.
2
.90
.85
.96
.9ft
.92
.90
.an
.87
.83
.8?
.86
.90
.85
.76
.73
.73
.7?
.nn
.84
.8(1
.83
. 83
.78
.81
.88
.84
.79
.84
.88
.85
84
Wind
"S c
3.!=
ax
05-n
11
26
74
76
24
77
26
25
75
76
24
76
75
74
75
75
25
74
?*>
23
74
25
24
24
24
25
25
24
25
d.
r F
S
14
4.7
6.2
8.6
8,0
6.2
3.9
8.6
4.9
8.5
4,8
8,2
10,0
7,5
4.9
5.2
6.7
7.2
6.4
5.7
5.5
5.5
7.3
6,8
4,9
6,5
6,4
6..1
8.9
7.8
6.9
6.6
,
&
B?.C
< E
15
8.8
8.6
9.6
8.6
8.2
7.1
9.2
7.6
8.6
6.2
8.5
10.5
8.9
7.6
6.9
7.8
7.5
7.9
6.8
7.9
7.B
8.3
7.3
7.5
8.9
8.9
7.8
9.2
9.6
7.5
8.2
Greatest in 24 hours and dates
Precipitation
of
Snow, ice pellets
0
Clear 9 Partly cloudy 14 Cloudy 7
Fastest
mile
^
in E
16
23
21
25
23
20
70
20
16
16
!•>
71
22
7?
70
17
1R
18
21
ID
IB
20
71
17
17
20
70
23
21
22
23
e
-2
w
a
17
sw/
w/
w/
sw/
sw/
sw/
w/
w/
w/
sw/
w/
w/
sw/
w/
w/
w/
H/
W/
w/
w/
sw/
w/
w/
w/
sw/
w/
w/
w/
w/
sw/
n .
251 W/
Sunshine
1
rf «
%-S
a c
353
18
£
•£ -3
0 O
fc*o
11
for
Sky cover
2
<£ ^
*" M
S c
c^S
20
6
9
6
g
10
10
9
7
3
4
0
2
4
5
6
0
2
4
0
7
7
9
9
4
3
2
gum
155
5.2
o
•fe-fe,
5 10' at Alaskan stations.
-t- Also on an earlier date, or dates.
X Heavy fog restricts visibility to '/4 mile or less.
T In the Hourly Precipitation table and in columns
9, 10, and 11 indicates an amount too small to
measure.
The season for degree days begins with July for heating
and with January for cooling.
Data in columns 6, 12, 13, 14, and 15 are based on 8
observations per day at 3-hour intervals.
Wind directions are those from which the wind blows.
Resultant wind is the vector sum of wind directions
9
10
11
12
1
2
3
P. M. Hour ending at
4
Subscription Price: Local Climatolog-
ical Data $ 1.00 per year including
annual summary if published. Single
copy: 10 cents for monthly summary;
15 cents for annual summary. Checks
or money orders should be made payable
and remittances and correspondence
should be sent to the Superintendent
of Documents, U. S . Goverment Print-
ing Office, Washington, D. C. 20402.
I certify that this is an official
publication of the National Oceanic
and speeds divided by the number of observations. and Atmospheric Admi
Figures for directions are tens of degrees from true is compiled from rec
North ;i.e., 09 = East, 18 = South, 27 = West, 36 = North, the National Climati
and 00 = Calm, When directions are in tens of degrees „?? 1 "a"onal , ,.„
in Col. 17, entries in Col. 16 are fastest observed VJ.J.J.B, n ^ i OQ_ i"1
1-minute speeds. If the / appears in Col. 17. speeds i^O r
are gusts. Lt J//, •il
Any errors detected will be corrected and changes in *^ VT/L*« — . / *••
summary data will be annotated in the annual summary. Director, National
nistration
ords on fi!
c Center, t
a 28801.
/„
^f^l
ClimaCic Ce
and
e at
kshe-
0
nter
5
6
7
8
9
10
11
12
£
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
75"
16
17
IB
19
20
21
22
23
24
25
26
27
28
29
30
SUMMARY BY HOURS
&
01
04
07
10
16
19
21
AVERAGES Resultant
, , . .— 1 u/mH
6
7
8
6
4
4
4
4
Station
pressure
In.
29.8:
29. 8j
29.8.
29.8-
29. 8(
29,8
29.8,
?9.8<
Temperature
61
60
62
67
> 69
) 68
. 63
r 62
33
3.0
59
58
59
62
63
62
60
59
>
Q
57
57
57
58
59
59
58
57
elative
umidity 7.
as:
88
89
85
74
69
72
82
86
1
!•= •-
1* -i
fe d
4.9 24
4.7 21
5.4 15
9.2 29
12.9 25
12.8 25
9.3 25
6.1 24
It
3.3
.8
1.7
8.9
12,7
12.6
9.3
5.4
USCOMM — NOAA — ASHEVILLE 400
-------
OBSERVATIONS AT 3-HOUR INTERVALS
a.
1
01
0*
07
10
13
16
19
22
01
04
07
10
13
It
19
22
04
07
16
19
22
01
0*
07
10
13
16
19
22
01
04
10
13
16
19
22
01
04
07
10
13
16
19
22
01
07
10
19
22
07
10
13
16
19
22
01
04
07
10
13
16
19
22
01
04
13
16
19
22
SKY COVER
0
0
e
9
2
3
5
8
0
0
0
0
0
2
3
0
10
10
3
6
4
10
10
10
10
10
10
3
10
7
10
1
2
9
0
0
0
9
2
1
10
5
2
7
6
10
8
0
0
10
3
1
0
0
0
10
10
10
10
1
4
10
10
10
10
0
0
0
0
DNI1H >
UNL
UNL
UNL
250
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
UNL
21
23
UNL
10
UNL
25
Ul.
S (
WEATHER
DA* 02
DAY 05
KH
DAY 08
DAY 11
DAY 14
KH
KH
KH
OAY 17
KH
KH
KH
KH
KH
KH
DAY 30
KH
KH
DAY 23
KH
DAY 26
KH
KH
KH
KH
DAY 29
DATA
ationa
icrof ic
and cc
ding, t
NGELE5
TEMPERATURE
b.
oe
<
56
56
57
60
67
66
59
57
57
64
66
60
59
63
61
61
61
6B
67
62
61
72
67
64
62
60
63
67
69
64
62
66
75
69
65
65
73
73
69
64
65
71
63
61
71
72
71
65
63
da
he,
sts
she\
CAL
It,
ii
54
53
54
55
58
58
55
53
56
60
57
59
58
58
58
62
59
60
66
62
61
59
62
63
63
61
60
63
63
62
66
67
62
62
65
61
59
64
60
59
as-
52
51
51
51
51
51
51
50
56
56
56
55
57
56
56
56
59
57
59
62
60
60
59
61
60
59
59
59
62
60
60
61
63
60
60
61
59
57
59
57
57
7
S
*#
ee
87
83
80
72
57
59
75
76
96
87
7S
70
87
81
84
B4
84
76
84
93
7J
87
93
97
93
78
73
71
64
90
87
66
73
84
66
71
87
84
71
87
87
64
75
81
WIND
I
19
16
06
24
24
!?
26
13
11
24
25
33
33
27
26
24
31
26
26
2<
27
19
22
24
22
2«
25
27
24
25
07
25
25
23
24
26
06
14
24
27
05
2»
26
2»
o«
p
6
5
4
a
12
11
9
5
4
5
6
U
12
4
3
9
8
8
5
11
12
10
2
12
4
3
3
5
8
9
10
8
6
4
4
13
7
6
B
IP
4
5
11
7
4
4
9
13
13
7
tt
SKY COV
10
10
9
9
2
4
6
0
10
10
10
4
7
10
10
10
10
10
10
0
0
0
0
0
0
4
10
10
i
1
5
S
4
0
9
0
1
5
3
10
4
10
10
2
10
0
0
0
0
0
0
CEILING
Hnds. (if fl.
42
33
45
45
UNL
UNI
uo
UNL
16
16
22
UNL
220
33
28
30
45
35
10
UNL
UNL
UNL
UNL
UNI.
UNL
UNL
8
12
UNI.
UNL
UNL
12
UNL
UNL
5
UNL
UNL
UNL
UNL
1*
UNL
11
UNL
13
UNL
UNL
UNL
UNL
UNL
UNL
VISI-
BILITY
fa
14
14
15
15
8
14
14
12
6
4
5
6
B
6
8
8
10
12
B
4
7
8
12
7
8
3
7
B
6
4
3
3
4
5
5
7
1
12
a
7
2
1
S
6
6
7
7
10
10
7
a
8
10
10
i*
6
8
DA* 03
D4Y 06
KH
KH
KH
KH
KH
KH
DAY 09
DAY 12
FK
OAY 15
KH
DAY 18
KH
KH
KH
KH
KH
KH
KH
DAY 21
FK
KH
DAY 24
KH
KH
KH
OAY 27
KH
KH
OAY 30
TEMPERATURE
u<
06
<
57
57
61
62
66
64
58
57
60
60
68
67
61
61
60
59
62
60
59
58
60
6B
69
62
61
62
70
61
62
61
63
70
70
6B
63
61
63
62
62
74
68
68
71
63
63
70
64
60
68
70
64
62
A
S3
53
55
59
59
Si
55
54
57
57
61
61
58
58
58
57
57
57
56
57
59
62
63
59
59
61
65
60
60
60
61
64
63
61
60
61
61
61
68
66
64
65
61
61
64
61
59
63
64
60
HE*.
a
a
50
50
50
50
54
J4
53
H
55
5i
56
56
56
56
56
56
54
54
54
56
58
58
5V
57
57
62
59
59
59
59
60
59
59
5*
60
61
62
61
62
59
6J
1»
61
59
57
61,
60
58
3
**
_1
a:
78
78
67
65
65
70
84
80
84
84
66
68
84
84
»7
90
75
81
64
93
93
71
73
84
87
»1
76
93
90
93
87
71
73
87
'3
90
97
66
78
73
67
90
Bl
73
64
90
76
71
81
WIND
te
0
26
26
12
25
25
25
26
25
12
04
15
24
25
27
18
26
24
26
25
26
24
23
23
25
25
26
24
19
25
25
24
26
18
08
26
24
26
25
00
07
24
26
26
20
10
17
23
25
24
25
23
27
or paper copies of Che original records. Inquiries as to
should be addressed to: Director, National Climatic Oncer
ille, North Carolina 28801.
FORNIA YEAR fe MONTH: 71 06
?J
6
5
4
13 NOTES
15
9
7 CEILING COLUMN—
UNL indicates an unlimited
2 ceiling.
5
7
12
1 j WEATHER COLUMN—
5
Tornado
T Thunderstorm
7 0 Squall
5 R Rain
9 RW Ram showers
10 ZL Freezing drizzle
V S Snow
4 SP Snow pellets
1C Ice crystals
SW Snow showers
5 SG Snow grains
4 IP Ice pellets
6 A Hail
1 1 1- Fog
u IF Ice fog
14 Cih Ground Tog
I? BD Blowing dusl
7 BN Blowing sand
BS Blowing snow
BY Blowing spray
3 K Smoke
u H Ha«
, D Dusl
10
10
10 WIND COLUMNS—
2 Directions urc those from
which the wind blows, indi-
cated in lens of degrees
-, from true North; i. e.. 09
f for East, IS for South. 27
f for Wesi, Entry of 00 in
, the direction column indi-
|° cates calm.
f Speed is expressed in knots;
5 multiply by 1.15 to convert
5 (o miles per hour.
0
8
a
10
3
10
tt
10
4
4
4
11
14
4
5
0
8
12
IQ
a
-124-
-------
APPENDIX B
PUBLIC ATTITUDE SURVEY ADMINISTRATION
Public attitude survey instructions to interviewers.
Public attitude survey questionnaire.
-125-
-------
PUBLIC OPINION SURVEY OF ODOR POLLUTION PROBLEMS
INSTRUCTIONS TO INTERVIEWERS
This survey is basically a study of public attitudes concerning odor problems as a form of
air pollution in selected areas of Los Angeles County, California. It is part of a study
being conducted by Copley International Corporation for the National Air Pollution Control
Administration. However, we do not wish the sponsor of the survey to be known as this
may tend to bias respondent answers .
It might be necessary in some cases to take definite steps to assure respondents that no
one will contact them to buy anything. Also, you might find it necessary to say that the
information they give becomes part of a statistical summary and individual responses will
not be exposed or revealed in any way.
What Numbers to Call. Each questionnaire will have a telephone number written on it.
This is the only number you should call. Numbers should not be substituted.
Make all the necessary notations in the spaces provided on the questionnaire for the results
to each call. Space is provided for recording data for an initial and five follow-up calls .
Just prior to starting the interview, write in the date and starting time. When you have
completed an interview, make a check mark under "Interview Completed." Make a check
mark under "Interview Not Completed" for each call for which an interview is not obtained.
Space is provided to record the best time to call in order to find the proper respondent at
home. This information may be obtained from other members of the household who answer
the call.
Space is also provided to record reasons why the interview was not completed, such as:
no answer, disconnected, refused, language problem, etc.
Whom to Interview. Interviews should be divided as evenly as possible between men and
women. Make alternate interviews with the male head of the household ("man of the house")
and the wife of the head of the household ("lady of the house"). Children, relatives, and
friends living in the household should not be interviewed. When there is only one head of
household, such as a single man or single woman, a widow, a divorcee, etc., then inter-
view that person.
Remember, when interviewing a household that has both a man and a woman available as
possible respondents, alternate your selection of respondents in order to maintain an
approximate 50-50 ratio between men and women.
-126-
-------
When to Call. Do not make any calls before 9:00 a.m. or after 9:30 p.m. unless you have
been instructed to call back at a specific time. Interviews will be conducted only on speci-
fied dates. In order to balance your quota by sex, you might have to make many follow-up
calls in the evenings. In making your follow-up calls, select a different time of day and,
where possible, an alternate (specified) interviewing day. In other words, spread out the
calls as much as possible to afford a greater opportunity of catching respondents at home.
General Questionnaire Instructions. Questions are to be asked exactly as they are written.
If a respondent indicates he does not understand a question, reread the question slowly.
Where a question is followed by three periods (for example, "Would you rate it...?"), read
the answers provided in the order they are listed, but do not read the answer numbers.
Never read the response "Don't Know." This is included for your convenience to check, if
appropriate.
Where a question does not have three periods at the end of it, do not read the responses.
In this case, simply check the number of the answer that corresponds to the respondent's
reply.
Check the number of the response given or write in the respondent's answer in the appro-
priate space. Do not mark anything on the lines numbered 2 through 73 in the extreme
right-hand margin of the paper.
Words and phrases which are underlined should be stressed. For instance, the major
portion of the questionnaire is concerned with the particular area of the city in which the
respondent lives, not the entire city as a whole. Therefore, in many questions the words
your area of the city are underlined and should be stressed.
The study is concerned with major odor pollution problems . Be sure the respondent does
not misunderstand your questions. We are not interested in odors produced from cooking
in his own kitchen, or from a neighbor who burns a pile of leaves once a year, or a neigh-
bor boy who has an old car with a smoky exhaust, etc. We are interested in odors
produced by factories, sewers, disposal areas, stockyards, large bodies of stagnant
water, and any other industrial or natural source which is a recognized irritation in the
selected areas of the County.
A telephone number will already have been written on the front page of each questionnaire.
After you finish an interview, fill in the respondent's name and address. Also write in
the name of the respondent's city (Hawthorne, Torrance, etc.). After the interview is
completed, sign your name on the last page.
Read all questions slowly and clearly, placing emphasis on words and phrases which are
underlined.
-127-
-------
Ql) After this question is read, read each of the five alternative answers (do not
read "Don't Know"), emphasizing each one and pausing between each one .
That is, read "Would you rate it.. .Excellent, Good, Fair, Poor, or Very Poor?"
Q2) Read the question as is, but do not read the responses. If the respondent
answers "Yes," then ask "What are they?" Do not read any of the answers
provided under this question. Check categories 1-4 only if these are specifi-
cally mentioned by the respondent. Write down any other answer or any answer
you are not sure how to categorize in the space provided next to "Other." If
the respondent mentions more than one thing his neighbors complain about,
indicate each of the things he says .
Q3A) On this question, read the three alternative answers slowly and carefully so
that the respondent can fix them in his mind. In some cases, it may be neces-
sary to repeat the alternative answers for the first two or three parts of the
question. For each problem, check one response only. For example, "Water
Pollution" may be answered either "Serious, " or "Somewhat Serious, " or "Not
Serious, " or "Don't Know, " but only one of these should be checked.
Q3B) If the respondent says there are other "serious" or "somewhat serious" prob-
lems, ask "What are they?" and record all that he says in the space provided.
Q4A & Do not read the responses to this question. If "Yes" is the answer, ask "What
Q4B) are they?" but do not read the following list of responses . Here again, write
down any other answer or any answer you are not sure how to categorize in the
space provided next to "Other." If the respondent mentions more than one con-
dition, indicate each of the things he says.
Q5A) Read the alternative answers slowly and carefully, repeating them if necessary.
For each item, check one response only. For example, "Smoke" may be
answered either "Yes, " or "No, " or "Don't Know, " but only one of these
responses should be checked.
Q5B) Do not read the responses at first. If "Yes" is the answer, ask "Which one is
the worst?" You may read the list of responses if the respondent does not
remember them or asks you to say them again. Check all the items that are
mentioned.
Q6A) Insert the name of the area of the city you are calling. Do not read the responses
to this question. If the answer is "No" or "Don't Know," skip parts B, C, D and
E.
-128-
-------
Q6B) Do not read the responses to this question. If "No" is the answer, skip part C
and ask parts D and E . If the answer is "Yes, " ask "How much would you say
it has bothered you? Is it. . .Very Much, Much, Moderately, Little, or Very
Little?" Do not read "Don't Know." Read the five alternatives slowly and
carefully . Repeat them if necessary . Check only one answer .
Q6C) Ask the question, including the responses . Read each response slowly and
carefully . Repeat them if necessary . Check only one answer .
Q6D) Ask the question as is . Do not read the responses . Before you begin inter-
viewing, you will be told the industry or type of plant(s) we are investigating.
If the respondent mentions this type of industry or a particular company engaged
in this type of processing, check "Plant being investigated." Any other type of
factory noise should be noted as "Other Factory." Accept only one answer to
this question, as we are interested in the source of most of the noise .
Q6E) Read this question with the responses . Read each response slowly and care-
fully, with a pause between each answer . Repeat if necessary. Check only
answer .
Q7A) Do not read the responses to this question. If the answer is "No" or "Don't
Know," skip parts B, C, D, E, F, G and H.
Q7B) Do not read the responses to this question . If the answer has to do with the
direction the wind is blowing, write the wind direction in on the line provided.
If the answer has to do with a particular process that a factory performs, write
the name of the process on the line provided. We are interested in when odors
are noticed most. Check only one answer .
Q7C) Ask this question and read the first four responses slowly and carefully. Do
not read "Don't Know." Repeat the answer if necessary or if the respondent
says something other than these four responses.
Q7D) Do not read the responses to this question. Check only one answer.
Q7E) Do not read the responses to this question. If "No" is the answer, skip part F
and ask parts G and H . If "Yes" is the answer, ask "How much would you say
they have bothered you? Is it. . .Very Much, Much, Moderately, Little, or
Very Little?" Do not read "Don't Know." Read these five alternatives slowly
and carefully . Repeat them if necessary . Check only one answer .
Q7F) Read the question and each response slowly and carefully. Do not read "Don't
Know . " Repeat the responses if necessary . Check only one answer .
-129-
-------
Q7G) Do not read the responses to this question. If the respondent mentions the
industry under investigation or a particular company engaged in this type of
processing, check "Plant being investigated." Any other type of factory odor
should be noted as "Other factory." Accept only one answer to this question.
Q7H) Read this question with the responses . Pause between each answer. Repeat
if necessary. Check only one answer .
Q8) Ask this question without the responses. If the answer is either "Good" or
"Bad," ask "Why is that?" In an open-ended question such as this, it is per-
missible and desirable to probe for additional answers, using a phrase such
as: "Can you think of any other reasons?"
Q9A) Do not read the responses to this question. If the answer is "No, " skip part B.
If the answer is "Yes," ask "Which problem?" You may clarify this question
by reading "Noise" or "Odor pollution" if necessary. If the respondent refuses
to tell you if he has contacted some authority, go on to part B.
Q9B) Do not read the responses to this question. The respondent may have contacted
more than one agency and, if so, check all of the ones he mentions.
Q10A) Do not read the responses . If the answer is "Yes, " ask "Which problem?"
You may read "Noise" or "Odor pollution" if necessary.
Q10B) Do not read the responses to this question.
QUA) Do not read the responses to this question. If the answer is "No" or "Don't
Know," skip parts B, C, D, E, F and G.
Q11B) Do not read the responses. If the answer is "No" or "Don't Know, " skip
parts C and D.
Q11C) Do not read the responses . Check as many answers as the respondent gives .
Q11D, Do not mark $0 unless the respondent actually says "Nothing" or "It hasn't"
Q11E, or something similar. If he cannot give you a figure, check "Don't Know/
Q11F) Refused."
Q11G) Do not read the responses to this question. Check only one answer.
Q12) Do not read the responses .
Q13) This is an open-ended question and probes such as "Can you think of any
others?" are desirable. Record any examples mentioned.
Q14) Do not read the responses .
-130-
-------
Q15) Do not read the responses. If the answer is "Yes, " ask "In what way?" and
probe for examples.
Q16) Ask this question and read the responses slowly and carefully. Repeat them
if necessary. Do not read "Don't Know." Check only one answer.
Q17) Do not mark $0 unless the respondent actually says "Nothing" or "I wouldn't
be willing to pay for it" or something similar. If he cannot give you a figure,
check "Don't Know."
Q18- These are personal data questions necessary for our statistical analyses.
Q23) Most people will readily give this information once the interviewer has estab-
lished rapport and the respondent feels assured the interview is legitimate.
However, there is always a small segment of the population who will refuse.
Do not pressure them to give this information if they are strongly inclined not
to do so.
Q19A) With this question, we are interested in the person's job title. If he does not
provide this information, you may probe further by asking such questions as
"What is your occupation?" or "What type of job is that?" Write in the job
title on the line provided, but do not check the appropriate category.
Q19B) Write in the industry on the line provided, but do not check the appropriate
category.
Finally, determine respondent's name and address and record this on the front page.
Check the sex on question 24, and terminate the interview by thanking the respondent for
his time and opinions. Be sure to sign your name on the last page.
-131-
-------
Mil
PUBLIC OPINION SURVEY OF
ODOR POLLUTION PROBLEMS
Respondent:
Phone No—
Address.
OMB No. 85-S70035
Approval Expires 6/30/71
Project No. (2-5 )
Area No. ( 6 )
Respondent No. (7-1O)
Record of Calls:
Call Date Time
1st
2nd
4th
nth
Interview
Completed
( )
( )
( )
( )
( )
Interview Not Best Time to Find Reason If
Completed Respondent at Home Not Completed
f
(
(
(
(
r ^
"Hello. My name is (name)
I'm calling (long distance) for a governmental agency interested in community problems. I'd like to talk with
the man/lady of the house to get his/her opinion on several questions." (ESTABLISH IF SPEAKER IS THE
RESPONDENT YOU SEEK. IF NOT. ASK IF RESPONDENT CAN BE CALLED TO THE PHONE.)
Ql) In general, how would you rate your area of the city as a place to live? Would you rate
it . . . ?
..Excellent
Poor
4
Good
-Very Poor
-Fair
11.
-Don't Know
5 - 6
SQ2) Can you think of any annoying things in your area that your neighbors complain about?
_No
_Yes (ASK: What are they?)
Don't Know 12
3
_Air pollution or odors
-Crime or area has deteriorated
Water pollution
3
Airport, industrial or traffic noise
13-
_Other_
-------
Q3A) Here are a few problems which different community areas are facing. How would you
rate each of these for your area of the city today in terms of being serious somewhat
serious, or not serious?
Water pollution
Noise in the community
Crime
Deterioration 01
utilities, public
Air pollution
Deterioration of stores, homes,
industrial buildings, etc.
i?
. Somewhat Not Don't
Serious Serious Serious Know
" is is— -TT-
1 ill
jnity
2 222
3 333
eets, parks,
lings, etc.
4 444
14
16
17
B) Are there any other problems that you think are serious or somewhat serious in your
community?
No _ Yes (ASK: What are they?)_ _ is_
*Q4A) Are there any conditions in your neighborhood that would cause you to spend your
leisure time inside your home rather than outside in the yard, neighborhood parks, etc.?
-No Yes (ASK: What are they?) Don't Know 20_
i 2 3
-Odors
1
-Noise
2
-Cold weather or wind 2i_
3
_Other_
*B) Are there any conditions in your neighborhood that would cause you to spend your
leisure time away from your home?
No Yes (ASK: What are they?) Don't Know 22.
12 3
-Odors
1
Noise
2
Cold weather or wind 23_
3
—Other.
Q5A) What do the words "air pollution" mean to you? Please answer "Yes" or "No" to the
following. Do they mean . . . ?
Yes No Don't Know
24 25 26
Smoke —=• r r
11 24.
Dust
Haze
26_
Noticeable odors — — -
Irritation of the eyes
Irritation of the nose or throat
-------
B) Are any of these a problem in your area of the city?
No Yes (ASK: Which one is the worst?) Don't Know 27_
12 3
Smoke
i
Dust
2
Haze
3
Fog
4 28_
Noticeable odors
5
Irritation of the eyes
6
Irritation of the nose or throat
Q6A) Some people here in (name of city) have been complaining recently
about noise. Have you noticed noise in your neighborhood during the last three
months?
-No (SKIP TO Q7A) __Yes (ASK FOLLOWING)
1 2 29_
Don't Know (SKIP TO Q7A)
a
B) Would you say that noise has bothered you?
No (SKIP TO Q6D) Yes (ASK: How much would you say 3O_
i 2 jt has bothered you? Is it. . . ?)
Very much Much
1 2
Moderately Little 3i_
-Very little Don't Know
s
C) How often has noise bothered you? Is it. . . ?
-Very often Often About half the time
12 3 32_
Infrequently Very infrequently Don't Know
45 6
D) Could you tell me where most of this noise comes from, that is, who or what causes it?
-Airplanes Plant being investigated
1 2
—Other factory . Traffic
-Other 33~
5
-Don't Know
6
E) In your area of the city, would you say that noise is . . . ?
Not a serious problem at all
-A more serious problem than it was a year ago
2
A less serious problem than it was a year ago 34-
A continuous and serious problem for over a year that has not
4 gotten worse or better
Q7A) Some people have also complained about odor pollution. Have you noticed any odors in
your neighborhood in the last three months?
.No (SKIP TO Q9A) Yes (ASK FOLLOWING)
1 2
—Don't Know (SKIP TO Q9A)
-------
*B) Have you noticed odors most during the morning, the afternoon, the evening or anv
other particular time? y
-During the morning
—During the afternoon
.During the evening
3
—When the air is still
4
—When the wind blows from the.
5 . __
When the plant does . 36-
All/most of the time
7
—Have not noticed it at a particular time
8
_ Other
9
-Don't Know
10
C) How often have you noticed these odors? Is it . . . ?
Everyday _ At least once a week
i
At least once a month ___ Only occasionally (less than once 37,
3 4a month)
-Don't Know
5
D) Generally speaking, how long do these odors last?
All the time
i
Two or three days
2
One day
More than one hour, but less than one day 38.
4
One hour or less
5
Other
7
.Don't Know
E) Would you say that these odors have bothered you?
_ No (SKIP TO Q7H) __ Yes (ASK: How much would you say they
i 2 have bothered you? Is it . . . ?) 39_
Very much Much
1 2
^Moderately Little
3 4 40.
.Very little Don't Know
5
F) How often have these odors bothered you? It is . . . ?
Very often Often About half the time
12 3 41
Infrequently Very infrequently Don't Know
5
G) What bothers you most about these odors? Would you say that it is . . . ?
- The number of times that you notice odors
.The strength of the odors in the air
2
42
The length of time that the odors last
t
3
Don't Know
-------
H) Could you tell me where most of these odors originate, that is, who or what causes
them?
-Plant being investigated
i
—Other factory
—Swamp/bay/river
3
Dump/incinerator/sewerage treatment
4 43.
Motor vehicle exhaust
5
—Other
6
-Don't Know
7
I) In your area of the city, would you say that odor pollution is . . . ?
Not a serious problem at all
-A more serious problem than it was a year ago
2
_A less serious problem than it was a year ago
A continuous and serious problem for over a year that has not
4gotten worse or better
Q8) Some people in your area of the city have mentioned that they buy bottled drinking
water. Let's suppose that you could also buy clean, odor-free air. How much would
you be willing to pay for odor-free air per month?
$ Don't Know 45.
Q9A) Do you own or are you purchasing this home?
No (SKIP TO Q10) Yes (ASK FOLLOWING)
1 2
-Don't Know/Refused (SKIP TO Q1O)
3
B) Would you tell me approximately what price you paid for your home?
$
-Don't Know Refused 47,
C) Do you feel that odor pollution has reduced the value of your home in any way, or not?
_No (SKIP TO Q10) Yes (ASK FOLLOWING)
i 2 48.
-Don't Know (SKIP TO Q1O)
3
*D) How has odor pollution reduced the value of your home?
Potential buyers are discouraged
.Guests/family members are uncomfortable
2
Causes deterioration of house
3
Can't use yard
4 49.
Don't go out
5
__Other
6
Don't Know
-------
E) In terms of dollars, how much would you estimate odor pollution has reduced the value
of your home?
Don't Know/Refused (ASK: In terms of a percentage, could you
estimate how much odor pollution has 50-
reduced the value of your home?)
% Don't Know/Refused 5i_
Q10) When houses in your neighborhood go up for sale, about how long does it usually take
to sell them?
_Less than a week _A week to a month
_Don't Know Refused
3 *
(IF "GOOD" OR "BAD," ASK: Why is that?)
Q13) Could you tell me how bad odors in a community might possibly affect the daily activi-
ties of the residents, if at all?
F) If your house was located in an area identical to yours but without the odor pollution,
how much more, if anything, do you estimate you would be able to get for your home if
you were to sell it?
$ Don't Know/Refused 52.
i 2
_One to three months . Four to six months
3 4
53-
—Seven to nine months Ten to twelve months
5 6
—More than a year Don't Know
7 8
Qll) If you consider the advantages and disadvantages for the people in your area of the
city, do you think it is good or bad to have industries that produce odors near your
neighborhood?
Good Bad
1 2 54.
55_
Q12) Have you ever seriously considered moving away from here because of odor pollution?
No _Yes Don't Know/Refused
123 56_
57.
.Don't Know
-------
Q14) Has a strong odor ever forced you or any member of your family to go indoors?
No Yes Don't Know/Refused
123
Q15) Do odors in the air ever make you or members of your family physically ill?
_No Yes .Don't Know/Refused
58_
123 59_
(IF "YES," ASK: In what way?)
60_
Q16A) Have you ever requested some authority or agency to take action concerning either
noise or odor pollution in your area of the city?
No (SKIP TO Q17) Yes (ASK: Which problem?) Refused
12 3 61.
Noise
i
_Odor pollution (ASK FOLLOWING)
2 62_
*B) What agency or authority did you contact?
Health Deoartment
63_
P.ity Hall
3
Other
5
Don't Know
2
Police Department
4
Q17A) Would you be willing to sign a complaint about either noise or odor pollution?
No ^ Yes (ASK: Which problem?) Don't Know
12 3 64 .
Noise
1
Odor pollution
65_
B) Would you be willing to appear in court to voice your complaint?
. No Yes Don't Know
1 2 3 66_
All of the information you've given us today will be kept strictly confidential. The following information is
needed to help us classify, analyze, and put data into summary report form. Individual replies will be kept
confidential and used only to calculate averages.
Q18) How long have you lived at your present address?
years_ months
67_
.Don't Know/Refused
-------
Q19A) What kind of work are you now doing?_
—Professional/technical Manager/official/proprietor
—Clerical /sales Skilled craftsman/foreman
—Operative/semi-skilled Service/non-skilled worker 68.
—Armed services Student
7 8
.Housewife _ Retired /unemployed
9 10
.Refused
11
B) (FOR CATEGORIES 1-6, ASK:)
In what kind of business or industry is this done?_
—Industry under investigation Other business
i 2
69
-Refused
3
Q20) Would you tell me approximately what age group you are in. Is it. . . ?
18-24 years 25-34 years _35-49 years
70_
50-64 years 65 years or over Refused
4 . 5 g
Q21) What is the highest grade of school you've completed?
Eighth grade or less Some high school
i 2
High school graduate Some college
71-
College graduate Post graduate
5 6
Vocational school _ Refused
7 8
Q22) How many persons, including yourself, live in this household?
One or two Three or four
1
Five or more —Refused
Q23) Considering your total yearly family income, what group would you say you're in? Is
it . . . ?
Less than $5,000 _$5,000-$7,999
_$8,OOO-$9,999 $10,OOO-$14,999
3 - 73.
_$15,OOO and over Don't Know
5
Refused
INTERVIEWER RECORD:
Q24) Sex:
Male Female
Interviewer's Name:.
-------
APPENDIX C
PUBLIC ATTITUDE SURVEY FINDINGS
Comparison of findings from public attitude surveys conducted in
TA(Hawthorne) and CA(Torrance).
-141-
-------
Ql. "In general, how would you rate your area of the city as a place to live? Would you rate it...?"
*>.
to
December
Response
Excellent
Good
Fair
Poor
Very poor
Don't know
All responses
Total no . of responses
Total no . of respondents
Q2 . "Can you think of any annoying things
TA(H)
26.5%
60.3
13.2
--
"
100.0%
68
68
in your area
CA(T)
26.0%
52.0
20.0
2.0
--
100.0%
50
50
that your
December a
Response
No
Yes
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
20.6%
77.9
1.5
100.0%
68
68
CA(T)
44.0%
56.0
--
100.0%
50
50
March
TA(H)
34.7%
45.3
16.0
4.0
—
100 .0%
75
75
CA(T)
34.0%
32.0
22.0
4.0
8.0
100.0%
50
50
June
TA(H)
32.0%
46.7
17.3
2.7
1.3
100.0%
75
75
neighbors complain about?"
March
TA(H)
46.7%
53.3
--
100.0%
75
75
CA(T)
40.0%
60.0
--
100 .0%
50
50
CA(T)
40.0%
53.3
4.0
2.7
--
100.0%
75
75
June0
TA(H)
33.3%
66.7
--
100.0%
75
75
CA(T)
50.7%
48.0
1.3
100.0%
75
75
aTA(H) and CA(T) are significantly different at the 1% level (X2 test).
bTA
-------
Q2 (Second Part). "What are they?"* (multiple responses) .
CO
December
Response
Air pollution or odors
Crime or area has deteriorated
Water pollution
Airport, industrial, or traffic noise
Dirt/litter
Drain/sewerage
Traffic
Refineries in area
Freeway in neighborhood
Public transportation
Schools
Drugs
High taxes
Other
All responses
Total no . of responses
Total no . of respondents
TA(H)
58.3%
2.8
4.2
22.2
--
1.4
--
--
2.8
--
--
--
--
8.3
100.0%
72
53
CA(T)
51.3%
2.9
34.3
--
--
2.9
--
--
--
--
--
--
--
8.6
100.0%
35
28
March
TA(H)
49.9%
--
4.2
22.9
2.1
--
2.1
--
2.1
--
--
--
--
16.7
100.0%
48
40
CA(T)
24.1%
--
6.1
27.3
3.0
--
6.1
27.3
--
--
--
--
--
6.1
100.0%
33
30
June
TA(H)
49.3%
--
--
14.8
4.9
--
1.6
--
9.8
--
1.6
—
1.6
16.4
100.0%
61
50
CA(T)
25.0%
--
--
22.7
6.8
2.3
13.6
4.5
--
--
--
--
--
25.1
100.0%
44
36
* Asked of the respondents who answered "Yes" to the previous question.
-------
Q3A. "Here are a few problems which different community areas are facing. How would you rate each of these
for your area of the city in terms of being serious, somewhat serious, or not serious?"
December
Response
Water pollution:
Serious
Somewhat serious
Not serious
Don't know
All responses
Total no . of responses
Total no . of respondents
Noise in the community:
Serious
Somewhat serious
Not serious
Don't know
All responses
Total no. of responses
Total no . of respondents
Crime:
Serious
Somewhat serious
Not serious
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
7.4%
8.8
80.9
2.9
100. 0%
68
68
5.9%
20.6
73.5
--
100.0%
68
68
2.9%
22.1
61.8
13.2
100.0%
68
68
CA(T)
28.0%
16.0
42.0
14.0
100.0%
50
50
18.0%
24.0
58.0
__
100 .0%
50
50
14.0%
26.0
54.0
6.0
100 .0%
50
50
Marcha
TA(H)
21.3%
17.3
56.0
5.4
100.0%
75
75
99 707
ZZ. //o
12.0
64.0
1.3
100.0%
75
75
Q W
7 .0/0
10.7
76.0
4.0
100.0%
75
75
CA(T)
20 .0%
16.0
46.0
18.0
100.0%
50
50
20.0%
16.0
64.0
--
100 .0%
50
50
14.0%
18.0
68.0
--
100.0%
50
50
Juneb
TA(H)
5.3%
22.7
60.0
12.0
100.0%
75
75
20.0%
28.0
52.0
--
100 .0%
75
75
13.3%
26.7
57.3
2.7
100.0%
75
75
CA(T)
25.3%
10.7
53.3
10.7
100.0%
75
75
13.3%
36.0
50.7
--
100.0%
75
75
22.7%
16.0
52.0
9.3
100.0%
75
75
-------
Q3A (Continued).
.
cn
December
Response
Deterioration of streets , parks , utilities ,
public buildings , etc . :
Serious
Somewhat serious
Not serious
Don't know
All responses
Total no . of responses
Total no . of respondents
Air pollution:
Serious
Somewhat serious
Not serious
Don't know
All responses
Total no . of responses
Total no . of respondents
Deterioration of stores, homes, industrial
buildings , etc . :
Serious
Somewhat serious
Not serious
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
1 W
1 .0/0
1.5
94.1
2.9
100.0%
68
68
38.2%
36.8
25.0
--
100.0%
68
68
1 SQ7
1 . O/o
13.2
79.4
5.9
100.0%
68
68
CA(T)
6.0%
8.0
86.0
--
100.0%
50
50
26.0%
40.0
32.0
2.0
100.0%
50
50
2.0%
8.0
90.0
--
100.0%
50
50
Marcha
TA(H)
2.7%
8.0
89.3
--
100.0%
75
75
49.3%
36.0
14.7
--
100.0%
75
75
5.3%
14.7
80.0
—
100.0%
75
75
CA(T)
2.0%
6.0
90.0
2.0
100.0%
50
50
44.0%
18.0
38.0
--
100.0%
50
50
10.0%
2.0
88.0
--
100.0%
50
50
Juneb
TA(H)
2.7%
4.0
92.0
1.3
100.0%
75
75
48.0%
34.7
17.3
--
100.0%
75
75
2.7%
4.0
93.3
--
100.0%
75
75
CA(T)
9 7
-------
Q3A (Continued).
aTA(H) and CA(T) are significantly different at the 1% level (X2 test).
bTA(H) and CA(T) are significantly different at the 5% level (X2 test).
Q4A. "Are there any conditions in your neighborhood that would cause you to spend your leisure time inside
your home rather than outside in the yard, neighborhood parks, etc.?"
December
Response
No
Yes
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
69.1%
30.9
--
100.0%
68
68
CA(T)
72.0%
28.0
--
100.0%
50
50
March
TA(H)
74.7%
25.3
--
100.0%
75
75
CA(T)
88.0%
8.0
4.0
100.0%
50
50
Junea
TA(H)
64.0%
34.7
1.3
100.0%
75
75
CA(T)
80.0%
18.7
1.3
100.0%
75
75
-TA(H) and CA(T) are significantly different at the 5% level (Xz test).
-------
Q4A (Second Part). "What are they?"* (multiple responses).
Response
Odors
Noise
Cold weather or wind
Smog
Smoke
Haze
Hard to breathe
Irritation of the eyes
Irritation of the nose
Irritation of the throat
Headaches
Area unsafe
Other
All responses
Total no . of responses
Total no . of respondents
* Asked of the respondents who answered "Yes"
aTA(H) and CA(T) are significantly different at
December
TA(H) CA(T)
61.7% 37.4%
7.7 12.5
15.4 31.3
3.8 6.3
3.8
3.8
__
-- --
--
__
__
3.8 12.5
100.0% 100.0%
26 16
22 10
March
TA(H)
50.0%
15.0
15.0
— fm
--
--
— •»
--
--
--
20.0
100.0%
20
19
CA(T)
38.4%
7.7
23.1
— —
--
--
7.7
--
--
23.1
-~
100.0%
13
12
June
TA(H)
54.9%
3.2
16.1
6.5
--
--
3.2
3.2
3.2
--
6.5
3.2
a
CA(T)
21.4%
7.1
57.3
7.1
7.1
__
•• —
--
--
__
--
100.0% 100.0%
31
26
14
14
to the previous question ,
the 10% level (X2 test) .
-------
Q4B. "Are there any conditions in your neighborhood that would cause you to spend your leisure time away
from your home?"
December
Response
No
Yes
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
89.7%
10.3
--
100.0%
68
68
CA(T)
88.0%
10.0
2.0
100 .0%
50
50
March
TA(H)
90.7%
9.3
--
100.0%
75
75
CA(T)
98.0%
2.0
—
100 .0%
50
50
June
TA(H)
90.7%
9.3
--
100.0%
75
75
CA(T)
89.3%
10.7
--
100.0%
75
75
oo
I
-------
Q4B (Second Part). "What are they?"* (multiple responses).
Response
December
TA(H) CA(T)
March
TA(H) CA(T)
June
TA(H) CA(T)
Odors
Noise
Cold weather or wind
Smog
Smoke
Haze
Hard to breathe
Irritation of the eyes
Irritation of the nose
Irritation of the throat
Headaches
Area unsafe
Other
All responses
66.7% 57.1% 75.0% 100.0% 100.0%
11.1 42.9 12.5
11.1
11.1
100.0% 100.0%
12.5
100.0% 100.0%
100.0
100.0% 100.0%
Total no. of responses
Total no. of respondents
9
7
7
5
7
7
8
8
* Asked of the respondents who answered "Yes" to the previous question.
-------
Q5A. "What do the words 'air pollution1 mean to you? Please answer 'Yes' or 'No' to the following. Do they
mean...?"
en
o
December
Response
Smoke:
Yes
No
Don't know
All responses
Total no . of responses
Total no . of respondents
Dust:
Yes
No
Don't know
All responses
Total no . of responses
Total no . of respondents
Haze:
Yes
No
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
52.9%
22.1
25.0
100.0%
68
68
33.8%
36.8
29.4
100 .0%
68
68
32.4%
42.7
24.9
100 .0%
68
68
CA(T)
70.0%
26.0
4.0
100.0%
50
50
56.0%
42.0
2.0
100.0%
50
50
58.0%
40.0
2.0
100.0%
50
50
March
TA(H)
62.7%
37.3
--
100.0%
75
75
62.7%
37.3
--
100.0%
75
75
69.3%
30.7
--
100.0%
75
75
CA(T)
84.0%
16.0
--
100 .0%
50
50
72.0%
28.0
--
100.0%
50
50
72.0%
22.0
6.0
100.0%
50
50
June
TA(H)
62.7%
37.3
--
100.0%
75
75
49.3%
49.3
1.4
100 .0%
75
75
73.3%
24.0
2.7
100.0%
75
75
CA(T)
69.3%
29.3
1.4
100.0%
75
75
62.7%
36.0
1.3
100.0%
75
75
49,3%
48.0
2.7
100.0%
75
75
-------
Q5A (Continued).
December
Response
Fog:
Yes
No
Don't know
All responses
Total no . of responses
Total no . of respondents
Noticeable odors:
Yes
No
Don't know
All responses
Total no . of responses
Total no . of respondents
Irritation of the eyes:
Yes
No
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
11.8%
57.4
30.8
100.0%
68
68
79.4%
8.8
11.8
100.0%
68
68
72.1%
20.6
7.3
100.0%
68
68
CA(T)
12.0%
86.0
2.0
100.0%
50
50
82.0%
16.0
2.0
100.0%
50
50
90.0%
8.0
2.0
100.0%
50
50
March
TA(H)
22.7%
76.0
1.3
100.0%
75
75
92.0%
8.0
--
100.0%
75
75
100.0%
--
--
100.0%
75
75
CA(T)
30.0%
70.0
--
100.0%
50
50
90.0%
8.0
2.0
100.0%
50
50
96.0%
2.0
2.0
100.0%
50
50
June
TA(H)
4.0%
93.3
2.7
100.0%
75
75
97.3%
2.7
--
100.0%
75
75
100.0
--
--
100.0%
75
75
CA(T)
12.0%
88.0
--
100.0%
75
75
97.3%
2.7
(
100.0%
75
75
98.7%
1.3
--
100.0%
75
75
-------
Q5A (Continued).
en
to
December
Response
Irritation of the nose or throat:
Yes
No
Don't know
All responses
Total no . of responses
Total no . of respondents
Q5B. "Are any of these a problem
TA(H)
69.1%
23.5
7.4
100.0%
68
68
in your area of the
CA(T)
84.0%
14.0
2.0
100.0%
50
50
city?"
December
Response
No
Yes
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
16.2%
82.4
1.4
100.0%
68
68
CA(T)
24.0%
76.0
--
100.0%
50
50
March
TA(H)
97.3%
2.7
--
100.0%
75
75
CA(T)
92.0%
4.0
4.0
100.0%
50
50
March3-
TA(H)
noq7
•3/0
82.7
--
100.0%
75
75
CA(T)
30.0%
66.0
4.0
100.0%
50
50
June
TA(H)
98.7%
1.3
--
100.0%
75
75
CA(T)
96.0%
4.0
__
100.0%
75
75
June*3
TA(H)
17.3%
81.3
1.4
100.0%
75
75
CA(T)
46.7%
53.3
__
100.0%
75
75
TA(H) and CA(T) are significantly different at the 10% level (X2 test).
bTA(H) and CA(T) are significantly different at the 1% level (X2 test).
-------
Q5B (Second Part) . "Which one is the worst?"*
Co
December
Response
Smoke
Dust
Haze
Fog
Noticeable odors
Irritation of the eyes
Irritation of the nose or throat
All responses
Total no . of responses
Total no . of respondents
* Asked of the respondents who answered
Q6A . "Some people here in (name
TA(H)
3.6%
3.6
1.8
57.1
10.7
23.2
100.0%
56
56
CA(T)
5.3%
7.9
7.9
10.5
39.4
21.1
7.9
100.0%
38
38
March
TA(H)
4.8%
1.6
6.5
62.9
16.1
8.1
100.0%
62
62
CA(T)
6.1%
9.1
3.0
18.2
33.3
30.3
100.0%
33
33
"Yes" to the previous question.
of city) have been complaining recently
June
TA(H)
1.6
3.2
1.6
65.6
23.0
5.0
100.0%
61
61
about noise .
CA(T)
2.5%
2.5
20.0
47.5
22.5
5.0
100.0%
40
40
Have
you noticed noise in your neighborhood during the last three months?"
December
Response
No
Yes
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
69.1%
30.9%
100.0%
68
68
CA(T)
60.0%
40.0%
100.0%
50
50
March
TA(H)
64.0%
36.0%
100.0%
75
75
CA(T)
60.0%
40.0%
100.0%
50
50
June
TA(H)
CA(T)
56.0% 64.0%
44.0% 36.0%
100.0% 100.0%
75
75
75
75
-------
Q6B. "Would you say that noise has bothered you?"*
en
£>.
i
December
Response
No
Yes
All responses
Total no . of responses
Total no . of respondents
* Asked of the respondents who answered
Q6B (Second Part) . "How much would
TA(H)
9.5%
90.5
100.0%
21
21
"Yes" to Q6A
you say it has
CA(T)
15.0%
85.0
100.0%
20
20
•
bothered you?
December
Response
Very much
Much
Moderately
Little
Very little
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
21.1%
21.1
36.7
15.8
5.3
--
100.0%
19
19
CA(T)
52.9%
5.9
35.3
5.9
--
--
100.0%
17
17
March
TA(H)
22.2%
77.8
100.0%
27
27
Is it...?"*
CA(T)
5.0%
95.0
100.0%
20
20
March
TA(H)
42.8%
14.3
23.8
--
19.1
--
100.0%
21
21
CA(T)
52.6%
15.8
15.8
15.8
--
--
100 .0%
19
19
June
TA(H)
42.4%
57.6
100.0%
33
33
CA(T)
33.3%
66.7
100.0%
27
27
June
TA(H)
52.6%
21.1
10.5
10.5
5.3
--
100.0%
19
19
CA(T)
27.8%
11.1
33.3
22.2
5.6
--
100.0%
18
18
* Asked of the respondents who answered "Yes" to the previous question.
-------
Q6C . "How often has noise bothered you? Is it.. .?"*
December
Response
Very often
Often
About half the time
Infrequently
Very infrequently
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
15.8%
26.3
15.8
26.3
10.5
5.3
100.0%
19
19
CA(T)
35.3%
17.6
11.8
29.4
5.9
100 .0%
17
17
March
TA(H)
19.1%
38.0
19.1
14.3
9.5
100.0%
21
21
CA(T)
26.3%
31.6
15.8
26.3
100.0%
19
19
June
TA(H)
57.8%
15.8
15.8
5.3
5.3
100.0%
19
19
CA(T)
16.7%
16.7
33.2
27.8
5.6
100.0%
18
18
1 Asked of the respondents who answered "Yes" to Q6B.
en
en
-------
Q6D. "Could you tell me where most of this noise comes from, that is, who or what causes it?"*
C/l
o\
December
Response
Airplanes
Plant being investigated
Other factory
Traffic
Construction
Animals
Trains
Motorcycles
Other
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
33.3%
57.1
--
--
--
4.8
4.8
100.0%
21
21
CA(T)
5.0%
35.0
45.0
--
--
--
15.0
--
100.0%
20
20
March
TA(H)
40.7%
40.8
--
--
3.7
14.8
--
100.0%
27
27
CA(T)
15.0%
35.0
35.0
5.0
—
--
5.0
5.0
100.0%
20
20
June
TA(H)
30.3%
45.5
--
--
3.0
18.2
3.0
--
100.0%
33
33
CA(T)
14.8%
14.8
55.6
3.7
--
--
7.4
3.7
--
100 .0%
27
27
-------
Q6E . "In your area of the city, would you say that noise is .. .?"*
01
December
Response
Not a serious problem at all
A more serious problem that it was a year ago
A less serious problem than it was a year ago
A continuous and serious problem for over a
year that has not gotten worse or better
All responses
Total no . of responses
Total no . of respondents
TA(H)
19.0%
52.4
28.6
100.0%
21
21
* Asked of the respondents who answered "Yes" to Q6A
Q7A . "Some people have also complained about odor
hood in the last three months?"
CA(T)
25.0%
40.0
35.0
100.0%
20
20
pollution .
December
Response
No
Yes
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
29.4%
70.6
100.0%
68
68
CA(T)
36 .0%
64.0
100.0%
50
50
March
TA(H)
18.5%
48.2
33.3
100.0%
27
27
CA(T)
20.0%
65.0
5.0
10.0
100.0%
20
20
Have you noticed any odors
Marcha
TA(H)
21.3%
7.8.7
100.0%
75
75
CA(T)
46.0%
54.0
100.0%
50
50
June
TA(H)
21.2%
54.6
24.2
100.0%
33
33
CA(T)
22.2%
48.2
7.4
22.2
100.0%
27
27
in your neighbor -
June*3
TA(H)
21.3%
78.7
100.0%
75
75
CA(T)
57.3%
42.7
100.0%
75
75
aTA(H) and CA(T) are significantly different at the 5% level (X2 test).
bTA(H) and CA(T) are significantly different at the 1% level (X2 test).
-------
en
oo
Q7B. "Have you noticed odors most during the morning, the afternoon, the evening, or any other particular
time?"* (multiple responses).
December
Response
During the morning
During the afternoon
During the evening
When the air is still
When the plant
All/most of the time
No particular time
When the wind blows from the W
When the wind blows from the N, NW
When the wind blows from the S, SW
When the wind blows from the E
Weekends
Warm weather
Other
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
1 7<7
L ' I/O
25.3
15.3
5.1
5.1
5.1
6.8
18.6
1.7
--
1.7
1.7
1.7
5.1
5.1
100.0%
59
48
CA(T)
33.2%
2.8
22.2
--
--
--
5.6
2.8
2.8
--
22.2
2.8
--
5.6
--
100.0%
36
32
March
TA(H)
UQ
-------
en
xO
I
Q7C. "How often have you noticed these odors? Is it.. .?"*
December
Response
Everyday
At least once a week
At least once a month
Only occasionally
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
12.5%
35.4
18.8
25.0
8.3
100.0%
48
48
CA(T)
__ Ct7
/O
50.0
25.0
25.0
100.0%
32
32
March
TA(H)
6.8%
35.6
32.2
23.7
1.7
100.0%
59
59
CA(T)
11.1%
29.6
29.6
7.4
22.3
100.0%
27
27
June
TA(H)
1.7%
35.6
23.7
37.3
1.7
100.0%
59
59
CA(T)
" %
21.9
31.3
43.8
3.0
100.0%
32
32
* Asked of the respondents who answered "Yes" to Q7A.
-------
Q7D. "Generally speaking, how long do these odors last?"*
December
Response
All the time
Two or three days
One day
More than one hour , but less than one day
One hour or less
As long as the wind blows
As long as the weather is right
Other
Don't know
All responses
Total no . of responses
Total no . of respondents
* Asked of the respondents who answered "Yes"
TA(H)
4.2%
4.2
10.4
54.2
6.3
6.3
2.1
6.3
6.0
100.0%
48
48
to Q7A .
CA(T)
" %
6.3
18.8
62.5
6.3
--
--
--
6.1
100.0%
32
32
March
TA(H)
" %
6.8
13.6
64.4
6.8
--
--
--
8.4
100.0%
59
59
CA(T)
" %
7.4
25.9
55.6
3.7
--
3.7
--
3.7
100.0%
27
27
June
TA(H)
1.7%
1.7
10.2
71.2
8.5
--
--
--
6.7
100.0%
59
59
CA(T)
" %
3.1
15.6
68.8
6.3
--
--
--
6.2
100.0%
32
32
-------
December
Response
No
Yes
All responses
Total no .
Total no .
of responses
of respondents
TA(H)
8.3%
91.7
100.0%
48
48
CA(T)
15.6%
84.4
100.0%
32
32
March a
TA(H)
22.0%
78.0
100.0%
59
59
CA(T)
3.7%
96.3
100.0%
27
27
June13
TA(H)
5.1%
94.9
100.0%
59
59
CA(T)
28.1%
71.9
100.0%
32
32
* Asked of the respondents who answered "Yes" to Q7A .
^TA(H) and CA(T) are significantly different at the 5% level (X2 test).
TA(H) and CA(T) are significantly different at the 1% level (X2 test).
Q7E (Second Part). "How much would you say they have bothered you? Is it...?"
December
Response
Very much
Much
Moderately
Little
Very little
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
43.1%
18.2
36.4
2.3
--
100 .0%
44
44
CA(T)
22.2%
7.4
48.2
3.7
18.5
100.0%
27
27
March
TA(H)
56.5%
15.2
23.9
4.4
--
100.0%
46
46
CA(T)
46.1%
19.2
30.8
3.9
--
100.0%
26
26
June
TA(H)
46.4%
19.6
25.0
3.6
5.4
100.0%
56
56
CA(T)
47.8%
13.0
34.8
4.4
--
100.0%
23
23
* Asked of the respondents who answered "Yes" to the previous question.
-------
Q7F. "How often have these odors bothered you? Is it.. .?"*
December
Response
Very often
Often
About half the time
Infrequently
Very infrequently
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
27.3%
25.0
11.4
29.5
4.5
2.3
100.0%
44
44
CA(T)
18.5%
25.9
18.5
25.9
11.2
100.0%
27
27
* Asked of the respondents who answered "Yes" to Q7E .
aTA(H) and CA(T) are significantly different at the 5% level (X2 test) .
Q7G. "What bothers you most about these odors? Would you say that
Response
The number of times that you notice odors
The strength of the odors in the air
The length of time that the odors last
Don't know
All responses
Total no . of responses
Total no . of respondents
* Asked of the respondents who answered "Yes"
December
TA(H)
11.4%
77.3
6.8
4.5
100.0%
44
44
to Q7E .
CA(T)
3.7%
88.9
7.4
100.0%
27
27
March
TA(H)
37.0%
39.1
8.. 7
10.9
2.2
2.1
100.0%
46
46
it is ...
CA{T)
15.4%
46.2
15.4
15.4
7.6
100.0%
26
26
?»*
March
TA(H)
4.4%
78.2
17.4
100.0%
46
46
CA(T)
3.9%
92.2
3.9
100.0%
26
26
June3
TA(H)
41.0%
25.0
16.1
12.5
5.4
100 .0%
56
56
CA(T)
30.4%
13.0
21.8
34.8
100.0%
23
23
June
TA(H)
1.8%
89.3
7.1
1.8
100.0%
56
56
CA(T)
4.4%
95.6
100.0%
23
23
-------
Os
Q7H. "Could you tell me where most of these odors originate, that is, who or what causes them?"*
December'
March0
* Asked of the respondents who answered "Yes" to Q7A.
aTA(H) and CA(T) are significantly different at the 1% level (X2 test).
Junec
Response
Plant being investigated
Other factory
Swamp/bay/river
Dump/incinerator/sewerage treatment
Motor vehicle exhaust
Other
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
89.6%
--
--
--
--
4.2
6.2
100.0%
48
48
CA(T)
" %
81.3
--
--
3.1
--
15.6
100.0%
32
32
TA(H)
79.7%
3.4
--
--
--
--
16.9
100.0%
59
59
CA(T)
" %
85.2
--
—
3.7
--
11.1
100 .0%
27
27
TA(H)
67.8%
1.7
--
3.4
--
1.7
25.4
100.0%
59
59
CA(T)
3.1%
59.4
—
--
--
--
37.5
100.0%
32
32
-------
Q7I. "In your area of the city, would you say that odor pollution is.. .?"*
December
Response
Not a serious problem
A more serious problem than it was a year ago
A less serious problem than it was a year ago
A continuous and serious problem for over a
year that has not gotten worse or better
All responses
Total no . of responses
Total no . of respondents
TA(H)
6.3%
54.1
4.2
35.4
100.0%
48
48
CA(T)
31.3%
37.4
31.3
100.0%
32
32
March
TA(H)
13.6%
30.5
8.5
47.4
100.0%
59
59
CA(T)
7.4%
55.6
3.7
33.3
100.0%
27
27
June
TA(H)
15.3%
17.0
18.6
49.1
100.0%
59
59
CA(T)
25.0%
18.8
3.1
53.1
100.0%
32
32
* Asked of the respondents who answered "Yes" to Q7A.
-------
Q8. "Some people in your area of the city have mentioned that they buy bottled drinking water. Let's suppose
C/l
that you could also buy clean, odor -free air. How much
would you be willing to
December
Response
$1 - $4
$5 - $8
$9 - $12
$13 - $19
$20 or more
Whatever it would cost
Nothing
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
12.5%
4.2
4.2
--
4.2
2.1
4.2
68.6
100.0%
48
48
CA(T)
3.1%
9.4
6.3
--
3.1
3.1
6.3
68.7
100.0%
32
32
pay for odor -free air per month?
March
TA(H)
3.4%
3.4
--
--
5.1
--
6.8
81.3
100.0%
59
59
CA(T)
" %
--
--
--
--
--
--
100.0
100.0%
27
27
June
TA(H)
" %
3.4
1.7
--
1.7
1.7
10.2
81.3
100.0%
59
59
CA(T)
3107
• i-7o
--
--
--
--
3.1
--
93.8
100.0%
32
32
; Asked of the respondents who answered "Yes" to Q7A.
-------
Q9A. "Do you own or are you purchasing this home?"
December
Response
No
Yes
Don't know/refused
All responses
Total no . of responses
Total no . of respondents
TA(H)
5.9%
92.7
1.4
100,0%
68
68
CA(T)
6.0%
94.0
--
100 .0%
50
50
March
TA(H)
8.0%
92.0
--
100.0%
75
75
CA(T)
8.0%
92.0
--
100.0%
50
50
June
TA(H)
4.0%
96.0
--
100.0%
75
75
CA(T)
8.0%
92.0
--
100.0%
75
75
o\
-------
Q9B. "Would you tell me approximately what price you paid for your home?"*
December
Response
Less than $5,000
$5,000 - $9,999
$10,000 - $14,999
$15,000 - $19,999
$20,000 - $24,999
$25,000 - $29,999
$30,000 - $34,999
$35,000 - $39,999
$40,000 - $44,999
$45,000 or more
Refused
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
" %
--
6.3
11.1
15.9
17.5
14.3
11.1
--
1.6
20.6
1.6
100.0%
63
63
CA(T)
" %
--
25.5
12.8
23.4
17.0
2.1
2.1
4.3
2.1
6.4
4.3
100.0%
47
47
March
TA(H)
" %
1.5
17.4
17.4
18.8
20.3
1.5
2.9
--
--
14.5
5.7
"100.0%
69
69
CA(T)
" %
2.2
15.2
15.2
13.0
21.7
2.2
2.2
--
--
17.4
10.9
100.0%
46
46
June
TA(H)
" %
--
13.9
4.2
15.3
22.2
11.1
4.2
--
--
23.6
5.5
100.0%
72
72
CA(T)
07
/O
__
33.3
24.6
15.9
7.3
4.4
—
--
1.5
10.1
2.9
100.0%
69
69
* Asked of the respondents who answered "Yes" to Q9A.
-------
o\
oo
I
Q9C. "Do you feel that odor pollution has reduced the value of your home in any way, or not?"
December
Response
No
Yes
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
79.4%
14.3
6.3
100 .0%
63
63
CA(T)
89.3%
6.4
4.3
100.0%
47
47
March
TA(H)
88.4%
8.7
2.9
100.0%
69
69
CA(T)
91.3%
2.2
6.5
100 .0%
46
46
June
TA(H)
84.7%
8.3
7.0
100.0%
72
72
CA(T)
94.2%
4.4
1.4
100.0%
69
69
* Asked of the respondents who answered "Yes" to Q9A.
-------
Q9D. "How has odor pollution reduced the value of your home?"* (multiple responses).
* Asked of the respondents who answered "Yes" to Q9C. Since so few respondents were asked this question,
statistical comparisons could not be made .
Q9E . "In terms of dollars, how much would you estimate odor pollution has reduced the value of your home?"*
* Asked of the respondents who answered "Yes" to Q9C. Since so few respondents were asked this question,
statistical comparisons could not be made.
4- Q9E (Second Part) . "In terms of a percentage, could you estimate how much odor pollution has reduced the
so value of your home?"*
* Asked of the respondents who answered "Don't know/refused" to the previous question. Since so few respon-
dents were asked this question, statistical comparisons could not be made.
Q9F. "If your house was located in an area identical to yours but without the odor pollution, how much more, if
anything, do you estimate you would be able to get for your home if you were to sell it?"*
* Asked of the respondents who answered "Yes" to Q9C. Since so few respondents were asked this question,
statistical comparisons could not be made,
-------
Q10. "When houses in your neighborhood go up for sale, about how long does it usually take to sell them?"
o
I
December
Response
Less than a week
A week to a month
One to three months
Four to six months
Seven to nine months
Ten to twelve months
More than a year
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
9 QQ7
Z.y/0
33.8
23.5
2.9
1.5
--
35.4
100 .0%
68
68
CA(T)
4.0%
32.0
34.0
8.0
--
--
22.0
100.0%
50
50
March
TA(H)
4.0%
49.3
20.0
--
--
--
26.7
100 .0%
75
75
CA(T)
16.0%
38.0
18.0
--
2.0
--
26.0
100.0%
50
50
June
TA(H)
5.3%
56.0
25.3
1.3
--
--
12.1
100.0%
75
75
CA(T)
2.7%
56.0
18.7
--
1.3
--
21.3
100.0%
75
75
-------
--4
i—i
I
Qll. "If you consider the advantages and disadvantages for the people in your area of the city, do you think it
is good or bad to have industries that produce odors near your neighborhood?"
December
Response
Good
Bad
Don't know
Refused
All responses
Total no . of responses
Total no . of respondents
TA(H)
5.9%
57.4
33.8
2.9
100.0%
68
68
CA(T)
2.0%
90.0
8.0
--
100.0%
50
50
Marcha
TA(H)
8.0%
76.0
16.0
--
100.0%
75
75
CA(T)
30.0%
68.0
2.0
--
100.0%
50
50
June a
TA(H)
1.3%
78.7
20.0
--
100.0%
75
75
CA(T)
13.3%
76.0
9.3
1.4
100.0%
75
75
aTA(H) and CA(T) are significantly different at the 1% level (X2 test).
-------
Qll (Second Part). Why is that?"* (multiple responses).
to
December
Response
Employment important/gives people jobs
Industry needed
Homes increase in value
Contributes to pollution/smog
Bad for general health of people
Hard to breathe
Annoying, objectionable
Unpleasant
Uncomfortable
Don't like to smell bad odors/want clean air
Should be separate from residential area/
need better zoning
Increases traffic
People move away
Have to close up house/windows/gets into
house
Have to stay indoors
Makes homes harder to sell
Decreases property values
Other
All responses
Total no . of responses
Total no . of respondents
TA(H)
O QQ7
Z.O/o
2.3
2.3
16.3
4.7
--
--
11.6
4.7
7.0
--
--
4.7
--
2.3
--
2.3
39.5
100.0%
43
25
CA(T)
3.6%
1.8
--
34.5
36.4
1.8
--
3.6
1.8
--
3.6
--
--
3.6
--
--
1.8
7.5
100 .0.%
55
42
March
TA(H)
1 W
L •1/o
9.9
1.1
16.5
13.2
5.5
9.9
6.6
5.5
6.6
5.5
2.2
--
1.1
2.2
2.2
3.3
7.6
100.0%
91
63
CA(T)
5.6%
14.8
--
22.2
3.7
1.9
1.9
5.6
5.6
25.9
11.1
--
--
--
--
--
--
1.7
100 .0%
54
49
June
TA(H)
2.0%
--
--
9.5
18.8
5.7
12.4
5.7
9.5
6.7
12.4
1.0
1.0
1.0
7.6
--
5.7
1.0
100.0%
105
59
CA(T)
2.5%
5.0
2.5
19.8
12.5
--
7.5
8.8
3.8
13.8
--
2.5
1.3
--
--
5.0
10.0
5.0
100.0%
80
67
* Asked of the respondents who answered "Good" or "Bad" to the previous question.
-------
Q12. "Have you ever seriously considered moving away from here because of odor pollution?"
December March June
Response
No
Yes
Don't know /refused
All responses
Total no. of responses
Total no . of respondents
TA(H)
77.9%
16.2
5.9
100.0%
68
68
CA(T)
80.0%
16.0
4.0
100.0%
50
50
TA(H)
78.7%
18.7
2.6
100.0%
75
75
CA(T)
66.0%
28.0
6.0
100.0%
50
50
TA(H)
70-307
/ O . O/o
22.7
4.0
100.0%
75
75
CA(T)
86.7%
13.3
--
100.0%
75
75
CO
-------
Q13. Could you tell me how bad odors in a community might possibly affect the daily activities of the residents,
if at all?" (multiple responses).
December
Response
Makes people stay indoors/don't use yard
People leave area for recreational activities
Causes headaches
Causes nausea
Affects breathing
Eye irritation/nose irritation/throat irritation
Slows people down
Feel uncomfortable
Affects general health
Coughing/ choking feeling
Causes allergies/makes them worse
Puts people in bad mood/makes you irritable
Gets into house/have to close up house
People move away
Wouldn't affect activities
There is none in this area
Other
Don't know
All responses
Total no . of responses
Total no. of respondents
TA(H)
6.9%
--
2.8
--
--
--
--
__
1.4
--
--
4.2
2.8
--
12.5
1.4
2.8
65.2
100.0%
72
68
CA(T)
32.7%
--
--
--
5.5
--
--
--
9.1
--
1.8
--
1.8
--
18.2
--
1.8
29.1
100.0%
55
50
March
TA(H)
29.2%
2.2
--
1.1
3.3
1.1
1.1
2.2
6.5
1.1
2.2
3.3
6.5
5.4
17.4
--
2.2
15.2
100.0%
92
75
CA(T)
2.0%
--
--
--
4.0
--
--
--
--
--
--
--
--
--
2.0
2.0
--
90.0
100,0%
50
50
June
TA(H)
34.9%
1.7
0.9
2.6
3.4
4.3
4.3
7.7
10.3
1.7
0.9
3.4
8.5
--
9.4
--
--
6.0
100.0%
117
75
CA(T)
41.0%
--
._
--
2.4
--
2.4
—
4.8
--
--
1.2
2.4
4.8
16.9
—
--
24.1
100.0%
83
75
-------
Q14. "Has a strong odor ever forced
Response
No
Yes
Don't know /refused
All responses
Total no . of responses
Total no . of respondents
December
TA(H) CA(T)
51.5% 64.0%
41.2 32.0
7.3 4.0
100.0% 100.0%
68 50
68 50
March
TA(H) CA(T)
54.7% 58.0%
45.3 40.0
2.0
100.0% 100.0%
75 50
75 50
Junea
TA(H)
45.3%
54.7
100.0%
75
75
CA(T)
74.7%
25.3
100.0%
75
75
aTA(H) and CA(T) are significantly different at the 1% level (X2 test).
i
i— -
C/l
1 Q15. "Do odors in the air ever make you or members of your family physically ill?"
Response
No
Yes
Don't know/refused
All responses
Total no . of responses
Total no . of respondents
December
TA(H) CA(T)
77.9% 74.0%
19.1 22.0
3.0 4.0
100.0% 100.0%
68 50
68 50
March
TA(H) CA(T)
70.7% 80.0%
28.0 20.0
1.3
100.0% 100.0%
75 50
75 50
Junea
TA(H)
72.0%
28.0
100.0%
75
75
CA(T)
88.0%
12.0
100.0%
75
75
aTA(H) and CA(T) are significantly different at the 5% level (X2 test).
-------
Q15 (Second Part). "In what way?"* (multiple responses).
December
Response
Nausea
Headaches
Eye irritation
Nose irritation
Throat irritation
Coughing/choking feeling
Causes allergies/makes them worse
Makes breathing difficult
Affects general health
Other
All responses
Total no . of responses
Total no . of respondents
TA(H)
33.2%
6.7
13.3
--
13.3
6.7
6.7
6.7
6.7
6,7
100.0%
15
13
CA(T)
26.7%
33.2
6.7
6.7
—
--
6.7
20.0
--
--
100 .0%
15
11
March
TA(H)
31.1%
19.2
11.5
3.8
3.8
3.8
7.7
11.5
3.8
3.8
100.0%
26
21
CA(T)
33.3%
8.3
—
16.7
--
--
16.7
16.7
8.3
—
100 .0%
12
10
June
TA(H)
28.6%
14.3
3.6
--
7.1
10.7
7.1
25.0
--
3.6
100 .0%
28
21
CA(T)
60.0%
10.0
--
10.0
--
--
--
20.0
--
--
100.0%
10
9
* Asked of the respondents who answered "Yes" to the previous question.
-------
Q16A. "Have you ever requested some authority or agency to take action concerning either noise or odor pollu-
tion in your area of the city?"
Decembera
Response
No
Yes
Refused
All responses
Total no . of responses
Total no . of respondents
a TA(H) and CA(T) are significantly different
Q16A (Second Part) . "Which problem?"*
TA(H)
69.1%
30.9
--
100.0%
68
68
CA(T)
88.0%
10.0
2.0
100.0%
50
50
March
TA(H)
74.7%
25.3
--
100.0%
75
75
CA(T)
84.0%
14.0
2.0
100.0%
50
50
June
TA(H)
80.0%
18.7
1.3
100.0%
75
75
CA(T)
Q1 QQ7
0 1 . O/o
18.7
--
100.0%
75
75
at the 1% level (X2 test) .
December
Response
Noise
Odor pollution
All responses
Total no . of responses
Total no . of respondents
TA(H)
9.5%
90.5
100.0%
21
21
CA(T)
60.0%
40.0
100 .0%
5
5
Marcha
TA(H)
5.3%
94.7
100.0%
19
19
CA(T)
28.6%
71.4
100.0%
7
7
June0
TA(H)
14.3%
85.7
100.0%
14
14
CA(T)
50.0%
50.0
100.0%
14
14
* Asked of the respondents who answered "Yes" to the previous question.
aTA(H) and CA(T) are significantly different at the 10% level (X2 test).
b TA(H) and CA(T) are significantly different at the 5% level (X2 test).
-------
Q16B. "What agency or authority did you contact?"* (multiple responses).
oo
I
December
Response
Air pollution control agency
Health Department
City Hall
Police Department
Citizens' groups
Neighborhood meetings
City government
State government
Federal government
Other
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
30.4%
8.9
8.9
8.9
8.9
21.7
4.3
--
--
4.3
3.7
100 .0%
23
21
CA(T)
" %
16.7
33.3
16.7
--
--
--
--
--
--
33.3
100.0%
6
5
March
TA(H) CA(T)
42.1% 71.4%
--
5.3 28.6
15.8
10.5
__
__
5.3
5.3
__
15.7
100.0% 100.0%
19 7
19 7
June
TA(H)
35.7%
--
--
14.3
7.1
21.4
--
--
--
7.1
14.4
100.0%
14
14
CA(T)
25.0%
--
18.8
31.3
--
--
--
6.3
--
12.3
6.3
100.0%
16
14
* Asked of the respondents who answered "Yes" to Q16A.
-------
Q17A. "Would you be willing to sign a complaint about either noise or odor pollution? "
December
Response
No
Yes
Don't know
All responses
Total no . of responses
Total no . of respondents
Q17A (Second Part) . "Which problem?"*
TA(H)
30.9%
41.2
27.9
100.0%
68
68
CA(T)
20.0%
58.0
22.0
100.0%
50
50
March
TA(H)
29.3%
57.3
13.4
100.0%
75
75
CA(T)
38.0%
62.0
100.0%
50
50
June
TA(H)
16.0%
58.7
25.3
100.0%
75
75
CA(T)
10.7%
84.0
5.3
100.0%
75
75
(multiple responses).
December3
Response
Noise
Odor pollution
All responses
Total no . of responses
Total no . of respondents
TA(H)
10.0%
90.0
100.0%
30
28
CA(T)
38.1%
61.9
100.0%
42
29
March
TA(H)
16.3%
83.7
100.0%
43
43
CA(T)
i ^ 9cy
lO.Z/o
84.8
100 .0%
33
31
June*5
TA(H)
13.6%
86.4
100.0%
44
44
CA(T)
34.9%
65.1
100.0%
63
63
* Asked of the respondents who answered "Yes" to the previous question.
aTA(H) and CA(T) are significantly different at the 1% level (X2 test).
bTA(H) and CA(T) are significantly different at the 5% level (X2 test).
-------
oo
o
Q17B. "Would you be willing to appear in court to voice your complaint?"
December3
Response
No
Yes
Don't know
All responses
Total no . of responses
Total no . of respondents
TA(H)
30.9%
26.5
42.6
100.0%
68
68
CA(T)
22.0%
44.0
34.0
100.0%
50
50
March3
TA(H)
38.7%
44.0
17.3
100.0%
75
75
CA(T)
66.0%
30.0
4.0
100 .0%
50
50
JuneD
TA(H)
26.7%
38.7
34.6
100.0%
75
75
CA(T)
9Q07
"Vo
80.0
10.7
100.0%
75
75
aTA(H) and CA
-------
APPENDIX D
TECHNICAL FIELD STUDIES DATA
Summary of percentage of time odor was detected and mean odor intensity
at panelist locations in TA(Hawthorne), TA(X Hawthorne), and TA(E1 Segundo).
Map of panelist locations in TA(Hawthorne) and TA(E1 Segundo) .
-181-
-------
ODOR INTENSITY RATING SUMMARY SHEET
Figure D-l. Summary of December ratings in TA(Hawthorne).
Panelist
Location
(Station)
142nd & Judah
141st & Glasgow
141st & Isis
140th & Tudah
139th & Glasgow
139th & Isis
139th & Judah
138th & Judah
137th & Glasgow
137th & Isis
135th & Glasgow
135th & Isis
135th & Judah
134th & Glasgow
134th & Isis
Wiseburn &
Hindrv
Wiseburn &
Tudah
Percent Time Detected and Mean Odor Intensity
Date:ll/30/7C
AM
%
M
C/5
w
b
fc
H
U
<
O
_1
w
k
9,
•*
1
i
3
H
Q
3
O
w
W
^
o
en
H
83
_,_}.
w
1
ttl
PM
%
M
Date:12/l/70
AM
%
M
CO
M
H
H
U
O
u
nJ
W
I— i
E
o
7;
g
s
s
H
Q
53
o
w
12-1
ff!
O
m
H
C/3
H-l
w
PH
PM
07
/O
M
Date: 12/2/70
AM
%
36
26
26
30
13
15
61
7
55
19
20
LOO
55
47
18
29
M
.4
.3
.1
.3
.1
.2
.7
.1
.7
.2
.3
1.4
.6
.7
.2
.3
PM
%
21
9
44
40
41
11
32
46
28
54
68
45
79
100
23
50
M
.1
.1
.5
.4
.4
.1
.3
.5
.5
.6
ID
.9
1.3
1.5
.2
.6
Date:12/3/70
AM
%
M
PM
%
39
0
0
9
0
2
20
19
20
26
2
8
56
20
0
14
M
.3
0
0
.1
0
0
2
.2
2
.3
0
.1
.6
.2
0
.1
Date:12/4/7C
AM
%
M
CO
2
0
H
3
w
dJ
BJ
W
U
tf
H
W
*_H
^H
g
•7;
W
U
W
J
w
Zj
Si
pcS
w
Q
O
^
O,
Id
S
EL)
Q
*?•
A
Q
W
-orr1
D
P
CO
3
1
-182-
PM
%
M
-------
ODOR INTENSITY RATING SUMMARY SHEET
Figure D-2. Summary of December ratings in TA(X Hawthorne).
Panelist
Location
(Station)
Imperial & Tudah
Imperial & Isis
Imperial &
La Cienegra
117th & Tudah
117th & Isis
117th &
La Cienega
119th & Tudah
119th & Isis
119th &
La Cienega
122nd & Judah
122nd & Hindry
124th & Judah
124th & Isis
124th & Glasgow
126th & Isis
126th & Glasgow
Percent Time Detected and Mean Odor Intensity
Date :i 1/30/70
AM
%
M
CO
M-4
P
i— i
•>
I-H
f-
$
o
-4
rvi
E
o
2
§
irf1
oej
H
a
<
O
&
z,
I-L)
M
&
Q
CO
H
C/2
3
w
2
^
PM
%
M
Date:i2/l/70
AM
17
/o
M
w
PM
P
1—4
*-»
1— 1
H
U
^j
D
_1
I-H
fe
o
fc
O
6
S
-(H
g
<^
O
S
2
U-l
M
Pi
rj
cn
H
V3
U
W
2
^
PM
%
M
Date: 12/2/70
AM
%
25
0
IOC
7
64
0
0
24
0
43
32
75
17
18
8
25
M
.3
0
1.3
.1
1.2
0
0
.2
0
.4
.6
.9
.2
.2
.1
.2
PM
<7
/o
0
0
0
4
7
0
0
4
0
0
0
9
0
0
0
3
M
0
0
0
0
.1
0
0
0
0
0
0
.1
0
0
0
0
Date: 12/3/70
AM
%
M
en
2
n
P
-------
ODOR INTENSITY RATING SUMMARY SHEET
Figure D-3. Summary of March ratings in TA(Hawthorne).
Panelist
Location
(Station)
142nd & Judah
141st & Isis
140th & Judah
138th & Judah
138th & Isis
138th & Glasgow
137th & Isis
135th & Judah
135th & Isis
135th & Glasgow
134th & Judah
134th & Isis
134th & Glasgow
Wiseburn &
Judah
Wiseburn &
Hansworth
132nd & Hindry
Percent Time Detected and Mean Odor Intensity |
Date:3/15/71
AM
%
M
w
Uj
s
p
o
"*$
Q
J
w
E
o
2
«\
P
U:
5
ttS
EH
13
z
-------
ODOR INTENSITY RATING SUMMARY SHEET
Figure D-4. Summary of March ratings in TA(E1 Segundo).
Panelist
Location
(Station)
Whiting &
Franklin
Virginia & Holly
Richmond &
Franklin
Richmond & Pine
Eucalyptus &
El Segundo
Eucalyptus &
Hollv
Penn &
El Segundo
Sierra & Hollv
Lomita &
Franklin
Maryland &
Grand
Holly &
Bungalow
Center &
Franklin
Oregon &
Franklin
Kansas &
Franklin
Illinois &
El Segundo
Percent Time Detected and Mean Odor Intensity
Date :3/15/71
AM
%
M
t/2
w
n
h— t
>
H
u
<
n
j
w
FT!
i
0
to
<
&
H
D
Z
-------
ODOR INTENSITY RATING SUMMARY SHEET
Figure D-5. Summary of June ratings in TA(Hawthorne).
Panelist
Location
(Station)
142nd & Judah
141st & Isis
140th & Judah
138th & Judah
138th & Isis
138th & Glasgow
137th & Isis
135th & Judah
135th & Isis
135th & Glasgow
134th & Judah
134th & Isis
134th & Glasgow
Wiseburn &
Judah
Wiseburn &
Hansworth
132nd &
Hindry
Percent Time Detected and Mean Odor Intensity
Date:6/14/71
AM
%
M
C/3
M-l
E
>-^
P
U
^
3
(— I
fe
Pi
Z
P
&
j
IS
i— i
(_>
I
w
w
rt
r -\
w
H
en
a
w
x
£
PM
%
M
Date:6/15/71
AM
%
8
16
38
36
41
16
51
66
56
51
6C
22
10(
8^
M
0
.1
.2
.3
.3
.1
.6
.9
.4
.5
.5
.3
1.7
1.3
PM
%
11
3
1
30
27
33
11
4
32
82
3
78
10
52
M
.1
0
0
.2
.1
.3
.1
0
.3
1.1
0
.4
.1
.5
Date:6/16/7
AM
%
1
10
1
1
LOO
0
15
11
28
48
28
25
29
M
0
.1
0
0
1.5
0
.1
.1
.2
.6
.2
.2
.2
PM
%
0
8
20
40
39
35
0
92
46
49
15
9
8
M
0
0
.1
.3
.3
.4
0
1.6
.6
.5
.2
.1
0
Date: 6/17/7
AM
%
89
0
0
0
5
52
12
32
37
23
8
9
1
7
M
.6
0
0
0
0
.6
.1
.4
.3
.1
0
0
0
0
PM
%
0
5
0
32
12
49
33
9
90
26
31
100
44
M
0
0
0
.5
.1
.5
.3
.1
.8
.3
.3
.6
.3
LDate;6/18//
AM
%
0
16
0
5
7
41
28
5
7
10
23
67
12
M
0
.1
0
.1
.1
.4
.2
0
.1
.1
.2
.8
.1
PM
Q7
/o
0
2
0
0
10(
33
19
17
10
34
12
40
27
M
0
0
0
0
1.2
.3
.1
.1
.1
.5
.9
.4
.2
-186-
-------
ODOR INTENSITY RATING SUMMARY SHEET
Figure D-6. Summary of June ratings in TA(E1 Segundo).
Panelist
Location
(Station)
Whiting &
Franklin
Virginia &
Holly
Richmond &
Franklin
Richmond & Pine
Eucalyptus &
El Segundo
Eucalyptus &
Holly
Penn &
El Segundo
Lomita &
Franklin
Maryland &
Grand
Bungalow &
Holly
Center &
Franklin
Kansas &
Franklin
Illinois &
Franklin
Indiana &
Grand
Percent Time Detected and Mean Odor Intensity
Date:6/14/71
AM
%
M
w
FT!
p
&
H
u
<1
9
rd
1— 1
PH
O
^H
f,
0
t=\
ri
E-
rK
2
ivi
w
&
r?
02
H
ft
J
W
z
£
PM
%
M
Date :6/ 15/71
AM
%
96
11
39
22
52
0
3
28
0
1
18
13
26
1
M
.6
.1
.4
.2
.3
0
0
.1
0
0
.1
.1
.2
0
PM
%
44
2
9
9
100
2
95
60
0
17
21
IOC
20
13
M
.6
0
0
0
.5
.4
1.5
.8
0
.1
.2
.8
.1
.1
Date:6/16/7:
AM
%
49
88
31
26
73
46
73
76
52
10
19
63
10(
M
.5
.6
.3
.2
1.1
.4
1.2
.9
.6
.1
.1
.5
1.7
PM
Q/
70
35
0
81
0
9
0
22
17
0
19
78
92
52
M
.3
0
.6
0
.1
0
.1
.1
0
.1
.8
.6
.5
Date<6/17/71
AM
%
55
4
19
28
56
8
55
45
28
20
46
28
22
29
M
.7
0
.2
.1
.5
.1
.8
.5
.2
.1
.4
.4
.2
.3
PM
%
0
14
41
12
46
18
38
33
28
0
0
61
90
0
M
0
.1
.4
.1
.5
.2
.3
.2
.2
0
0
.4
.6
0
Date £/l 8/7
AM
%
57
0
25
0
50
50
43
13
9
22
44
50
4
M
.5
0
.1
0
.7
.8
.6
.1
.1
.1
.5
.5
.1
PM
%
91
0
60
9
25
56
12
13
0
17
6
37
20
M
IX)
0
5
.1
.3
.7
.2
.1
0
ri
.^
0
c
• I
,1
-187-
-------
Figure D-7. Percentage of time odor was detected and mean odor intensity at
panelist locations in TA(Hawthorne) during March 1971 technical field studies.
To Oil Refinery
-188-
-------
1,000'
P I—
fD C
w'
rf
H- O
O «
o <»
P •
Mariposa Ave.
I
k—
CX3
I
P
CTQ
ro
O
Sa
o
8- 4
at) m
2 Cu
03 r^
hj fD
Oil Refinery Compound
rt &
fD H
o 3
D* fD
P P
r^
cn 0)
, .
r*- m-
fD v<
." P
-------
Figure D-9. Percentage of time odor was detected and mean odor intensity at
panelist locations in TA(Hawthorne) during June 1971 technical field studies.
To Oil Refinery
-190-
-------
0
1,000'
CD C
£t i-t
w o>
o
S
Mariposa Ave.
CO
cri
M
•T-l
oj
a
•
fcuO
1
^
^
•
CO
cd
•*H
~ d
•Et>
£
>
•
53%
0.5
•M
CO
"B
Q
d
o
_U
0.1
T3
d
-1
,£5
O
•H
c£
Grand
• l3%
0.1
<*^—
Ave
•38%
0.3
CO
-a
-a
T3
i
4->
CO
l^>
Frai
El Segundo Blvd.
Q
3
-ft
o
M
^^
iklin
g
T3
Q)
CO
0.1
*>
Ave
51%
0.5
A^
g
/
s
tr
CO
*
a
fi
/
•
01
f_l
49% VH
• CO
iTl
1
1
Holb
J
^
s
o
J
.
co
TD
I
iJ
39W^
° in
i
^
[
•
i
N,
Av
9%
0.1
+J
CO
^ n
cj CD
of*
31 w
mlu
V
X
e.
%
27%
0.2
N
">
4-1
CO
Clj
• T-l
. G
o
<4H
• r-l
r— )
oi
o
Pine
N.
W
w
^
1
s^
"V
0
52%
0.4
Ave
J
*s
•iH
O
,g
i — i
i-H
1 — (
i
s&
A tvi
49°§
g
rf
>
• ^
11 °{
w
H3
5
iff
E' o
3 CD
^ o
CO "^
II
3 CD
a, _
a.
h— ft)
^ 5
g "
CD n
Oil Refinery Compound
e. 2
O
CD
g
-------
APPENDIX E
EFFECTS DETERMINATION QUESTIONNAIRE
-193-
-------
PUBLIC ATTITUDE SURVEY OF COMMUNITY ODOR PROBLEMS
EFFECTS DETERMINATION QUESTIONNAIRE
Respondent:
Phone No.
Street Address
City
File No.
Survey No.
Respondent No.
Record of Call:
Date
Time
Check If
Interview
Completed
Check If
Interview Not
Completed
Reason If
Not Completed
(name)
. I'm calling (long dis-
"Hello. My name is
tance for a government agency interested in certain community problems.
I'd like to talk with the man/lady of the house to get his/her opinion on a few
questions. Are you the man/lady of the house?" (IF NOT, ASK: "May I talk
with him/her?")
Ql) Have you lived at your present address more than six months?
1. Yes
2. No (SAY: "I'm sorry. This survey was designed for
longer term residents . Thank you for your time.")
Q2) Lately, people have become very much concerned about the environ-
ment and various types of pollution. Do people complain about any
kind of pollution in your neighborhood?
1. Yes(ASKQ3.)
2. No(ASKQ4.)
3. Don't know (ASK Q4.)
Page 1
-------
Q3) Do people complain about...?
1. Air pollution in your neighborhood
2. Water pollution in your neighborhood
3. Airport, industrial, or traffic noise in your neighborhood
4. Noticeable odors in your neighborhood
5. Don't know (DO NOT READ THIS RESPONSE.)
Q4) Have you noticed any odors in your neighborhood in the last three
months?
1. Yes (ASK Q5.)
2. No (SKIP TO Q23.)
3. Don't know (SKIP TO Q23.)
Q5) How often have you noticed these odors?
1. At least once a day
2. At least once a week, but not every day
3. At least once a month, but not every week
4. Less than once a month
5. Don't know
Q6) Generally speaking, how long do these odors last?
1. At least one day
2. At least one hour, but less than one day
3. At least 15 minutes, but less than one hour
4. Less than 15 minutes
5. Don't know
Page 2
-------
Q7) How strong would you say these odors smell? Would you say...?
1. Very strong
2. Strong
3. Moderate
4. Slight
5. Don't know (DO NOT READ THIS RESPONSE .)
Q8) When was the last^ time you noticed odors in your neighborhood?
1. This week
2. Last week
3. Two to four weeks ago/a month ago
4. More than a month ago
5. Don't know
Q9) Would you say that these odors have bothered you?
1. Yes(ASKQ10.)
2. No (SKIP TO Qll.)
3. Don't know (SKIP TO Q11.)
Q10) How much would you say they have bothered you? Would you say...?
1. Very much
2. Much
3. Moderately 10
4. Little
5. Don't know (DO NOT READ THIS RESPONSE .)
Page 3
-------
Qll) Has a strong odor ever forced you or any member of your household
to go indoors?
1. Yes(ASKQ12.)
2. No(SKIPTOQ13.) H
3. Don't know (SKIP TO Q13.)
Q12) When was that?
1.
12
2. Don't know
Q13) Has a strong odor ever forced you or any member of your household
to temporarily leave your neighborhood?
1. Yes(ASKQ14.)
2. No (SKIP TO Q15.) 13_
3. Don't know (SKIP TO Q15.)
Q14) When was that?
1.
14
2. Don't know
Q15) Have odors in your neighborhood ever made you or members of your
household physically ill?
1. Yes(ASKQ16.)
2. No (SKIP TO Q19.) 15_
3. Don't know (SKIP TO Q19.)
Page 4
-------
Q16) In what way?
1. Nausea
2. He adache s
3. Affected breathing/caused coughing or choking 16
4. Other answer(s):
5. Don't know
Q17) Has any member of your household visited a doctor because of these
illnesses?
1. Yes(ASKQ18.)
2. No (SKIP TO Q19.) 17_
3. Don't know (SKIP TO Q19 .)
Q18) When was that?
1.
18
2. Don't know
Q19) How much are homes in your neighborhood selling for lately?
1. dollars
19
2. Don't know
Q20) Have odors reduced the value of homes in your neighborhood?
1. Yes(ASKQ21.)
2. No (SKIP TO Q22.) 20
3. Don't know (SKIP TO Q22.)
Page 5
-------
Q21) In what way?
1. Potential buyers are discouraged
2. Owners are anxious to move away
21
3. Other answer(s):
4. Don't know
Q22) Where would you say most of the odors in your neighborhood originate,
that is, who or what causes them?
1. Source(s) being investigated
2. Other source(s): 22
3. Don't know
"All this information is strictly confidential, but we need it for statistical
purposes."
Q23) What company do you work for?
1. Source(s) being investigated
_2. Other company/none: 23
3. Refused
Q24) What companies do other members of your household work for?
1. Source(s) being investigated
2. Other company(ies)/none: 24
3. Refused
"That completes the interview. Thank you very much for your time."
Interviewer's Name:
Page 6
-------
APPENDIX F
GLOSSARY OF TERMS
-201-
-------
APPENDIX F
GLOSSARY OF TERMS
Acute Episode
Chronic Situation
Dose/Response
Economic Effects
Noise
Odor
Odorant
Odorous Emission
Odor Judgment Panel
Causation of odors beyond the property limits of a
source, occurring no more than once and lasting no
more than one day in any three month period.
Causation of odors beyond the property limits of a
source, occurring more than once or lasting more
than one day in any three month period.
The degree and type of human response caused by
variations in exposure to different intensities and/or
other characteristics of odors.
Consequences of odors that can be measured in mone-
tary terms, e.g., reduced value of real estate, in-
creased maintenance costs, and lost of work productivity.
Responses based on the interaction of variables other
than the one of concern to the research team; speci-
fically, responses by individuals that they noticed or
were bothered by odors in their neighborhoods when
the location of sources or meteorological conditions
were such that odors could not have been detected.
(1) Perception of smell, referring to the experience;
(2) an odor ant.
That which is smelled, referring to the stimulus; a
substance that stimulates the olfactory receptors.
Release of an odorant, or a combination of odorants
that can be smelled, into the atmosphere.
A group of individuals selected on the basis of their
sensitivity to odorants, the reliability of their evalu-
ation of odors, and their having socioeconomic char-
acteristics similar to the population of a community
under study as an odor problem area. The panel is
used to obtain information concerning the subjective
attributes of odors perceived in that community.
-202-
-------
Odor Plume
Reference Standard
Scentometer
Signal
Social Effects
Triangle Test
A three dimensional, featherlike pattern of an odorant,
or a combination of odorants that can be smelled, ema-
nating and diffusing from a point source.
A laboratory prepared sample of a non-toxic, stable
chemical that closely matches the odor under investi -
gation and which, when mixed to various concentrations,
is used to train odor judgment panel participants to
recognize the strengths of the odor perceived during
field investigations.
A device for odor measurement by sensory techniques
that employs vapor dilution methods in which the odor
is diluted with increasing volumes of odorless air
until it is just discernible.
Responses based on the interaction of the variable of
concern to the research team, specifically responses
by individuals that they noticed or were bothered by
odors when odors were actually present.
Consequences of odors that can be measured only in
subjective or medical terms; specifically, illness
and changes in preferred life style, e.g., being
forced to stay indoors because of odors .
A psychophysical method for determining whether a
difference between two odorants can be detected, in
which individuals are asked to choose which of three
odorants is the odd or different one, where two are
alike and one is different.
-203-
-------
ANNOTATED BIBLIOGRAPHY
-205-
-------
ANNOTATED BIBLIOGRAPHY
A. EFFECTS OF ODOR POLLUTION ON THE COMMUNITY
Deane, Margaret, and Goldsmith, JohnR. "Health Effects and Annoyance
Reactions to Pulp Mill Odor in a Rural Community." Paper presented
at the working conference on The Dose-Response Relationships Affect-
ing Human Reactions to Odorous Compounds, a symposium sponsored
by the Environmental Protection Agency, Air Pollution Control Office,
Cambridge, Massachusetts, April 26-27, 1971.
Results of a study conducted in a rural community exposed to odors
from a pulp mill are presented. Three levels of exposure were tested
and compared.
DeGroot, Ido. "Trends in Public Attitudes Toward Air Pollution, " Journal of the
Air Pollution Control Association, XVII, No. 10 (October, 1967), 679-681,
Findings relative to people's attitudes regarding the air pollution prob-
lem from a series of studies conducted by the U.S. Public Health
Service are reported, and trends are noted.
Engen, Trygg. "Method and Theory of Odor Preference." Paper presented at
the working conference on The Dose-Response Relationships Affecting
Human Reactions to Odorous Compounds, a symposium sponsored by the
Environmental Protection Agency, Air Pollution Control Office,
Cambridge, Massachusetts, April 26-27, 1971.
This paper takes a developmental view of the hedonics of odor percep-
tion in considering the general problem of method and theory. Results
of studies with adults and children are discussed.
-207-
-------
Flesh, R. David. "Property Value Differentials as a Measure of Economic
Costs Due to Odors." Paper presented at the working conference on
The Dose-Response Relationships Affecting Human Reactions to Odor-
ous Compounds, a symposium sponsored by the Environmental
Protection Agency, Air Pollution Control Off ice, Cambridge,
Massachusetts, April 26-27, 1971.
Two methods of measuring property value differentials due to odors —
the "other things equal" approach and the problem rating scale — are
presented and discussed. Preliminary findings of a present study of
the first method are presented.
. "Public Opinion of Odors in Seven Metropolitan Areas ." Paper pre-
sented at the Conference on Methods for Measuring and Evaluating
Odorous Air Pollutants at the Source and in the Ambient Air, Stockholm,
June 1-5, 1970.
This paper discusses a study which surveyed all local air pollution con-
trol agencies in the nation and public opinion of odors in seven metro-
politan areas. The paper describes the development of the public opinion
survey and summarizes the results.
Foster, Dean. "Working Paper for Stockholm Methods Conference." Paper
presented at the Conference on Methods for Measuring and Evaluating
Odorous Air Pollutants at the Source and in the Ambient Air, Stockholm,
June 1-5, 1970.
This paper discusses the lack of previous attention devoted to the public
health importance of odor because of difficulties regarding scientific
measurement and the lack of easy verbalization of response patterns.
The author calls for a better use of resources to define the problem.
Friberg, Lars, Jonsson, Erland, Rylander, Ragnar, and Ubisch, HansV.
"Reactions to Environmental Disturbance Factors Caused by Oil Refin-
eries." Institute of Hygiene Karolinska Institutet, Department of
General Hygiene, National Institute of Public Health, Stockholm, and
Institute of Hygiene Umea University, Umea, undated.
A discussion of a prognosis made for the expected annoyance reaction
in the population around a new refinery is presented. The prognosis
was based on a technical investigation, literature studies, and an inter-
view survey.
-208-
-------
Goldsmith, John R. "A Suggested Odor Scaling System." Paper presented at
the Conference on Methods for Measuring and Evaluating Odorous Air
Pollutants at the Source and in the Ambient Air, Stockholm, June 1-5,
1970.
This paper prepares a logarithmic scale which establishes the relation-
ship between odorant presence, odor-dilution panel study responses, and
the extent to which populations are bothered by odor exposure.
Jonsson, Erland. "On Annoyance Reactions Observed Within Human Populations ."
Paper presented at the working conference on The Dose Response Rela-
tionships Affecting Human Reactions to Odorous Compounds, a symposium
sponsored by the Environmental Protection Agency, Air Pollution Control
Office, Cambridge, Massachusetts, April 26-27, 1971.
The effects of offensive odors on the exposed population as reported in a
number of investigations are reviewed.
, Deane, Margaret, and Sanders, George. "Community Reactions to
Odors from Pulp Mills: A Pilot Study in Eureka, California." Paper
presented at the Conference on Methods for Measuring and Evaluating
Odorous Air Pollutants at the Source and in the Ambient Air, Stockholm,
June 1-5, 1970.
The purpose of this study was to determine whether or not it is possible
to measure the relationship between quantitative data describing exposure
and quantitative data describing community reactions. The results of the
study showed a definite association between exposure data and reaction
data.
Kigo, Tetsutaro, Endo, Ryosaku, Tatsunori, Oyake, and Shirakawa, Hiroshi.
"Offensive Odors. 9 . Investigation of the Existing Conditions in
Hokkaido With Respect to Offensive Odors." First Report. Taiki Osen
Kenkyu (J. Japan. Soc. Air Pollution), II, No. 1 (1967), 48-58 (Trans-
lation).
This article is a report of an investigation of the sources of odors,
damage being caused, and consequent reactions and attitudes of local
residents in Hokkaido.
-209-
-------
Pavenello, R., and Rondia, D. "Odour Nuisance and Public Health." Paper pre-
sented at the Conference on Methods for Measuring and Evaluating
Odorous Air Pollutants at the Source and in the Ambient Air, Stockholm,
June 1-5, 1970,
This paper reviews some of the technical problems associated with odor
pollution as well as public health prpblems, reactions, and prevention of
pollution by odoriferous products.
Sadilova, M.S. "Studies in the Standardization of Maximum Allowable Hydrogen
Fluoride Concentrations in the Air of Inhabited Areas," in The Biological
Effects and Hygienic Importance of Atmospheric Pollutants , Book 10
(U.S.S.R. Literature on Air Pollution and Related Occupational Diseases,
1968), XVII, 118-128.
An investigation of the levels of atmospheric fluorine found near alumi-
num and cryolite plants, accompanied by a survey of the health of
children who live in these areas has shown that the establishment of the
maximum allowable concentration of fluorine is imperative. Odor per-
ception thresholds and physical effects of HF on warm-blooded animals
are presented.
Springer, Karl J. "Studies of Public Opinion of a Traffic Odor," Paper presented
at the working conference on The Dose-Response Relationships Affecting
Human Reactions to Odorous Compounds, a symposium sponsored by the
Environmental Protection Agency, Air Pollution Control Office, Cambridge,
Massachusetts, April 26-27, 1971.
This paper describes an effort to acquire public opinion of diesel exhaust
in terms of the Turk kit panel ratings of the same odor. The intent of
the study was to relate and close the loop between technology, measure-
ment, and response.
, and Hare, Charles T. "A Field Survey to Determine Public Opinion
of Diesel Engine Exhaust Odor." Final Report. Southwest Research
Institute, San Antonio, Texas, Vehicle Emissions Research Laboratory,
NAPCA Contract PH 22-68-36, February, 1970.
This study developed a workable method of sampling public opinion of
diesel exhaust odor by using a calibrated mobile odor evaluation labora-
tory. Five cities were surveyed. Unsolicited comments made during
the survey seemed to indicate that high local pollution and/or high aware •
ness of it are associated with a more objectionable characterization of
odors, and vice versa.
-210-
-------
Summer, W. "Odor Research in Europe: A Report, " Annals of the New York
Academy of Sciences, CXVI, Art. 2 (July, 1964), 361-368.
The approach to the practical problems of odor research in Europe
differs from that in the United States. The author provides a picture
of research in Europe and the implications of and reasons for much of
the work that is being carried out.
-211-
-------
B. EFFECTS OF ODOR POLLUTION ON THE INDIVIDUAL
Altner, H. "Neurological Responses to Odorants." Paper presented at the work-
ing conference on The Dose-Response Relationships Affecting Human
Reactions to Odorous Compounds, a symposium sponsored by the Environ-
mental Protection Agency, Air Pollution Control Office, Cambridge,
Massachusetts, April 26-27, 1971.
A brief survey of the neurological responses to odorants of some animals
is provided. Topics covered are the input to the central nervous system,
pathways in the olfactory centers of mammals, electrophysio logical in-
vestigation of the olfactory bulb and related forebrain structures .
Amoore, John E. "Current Status of the Steric Theory of Odor, " Annals of the
New York Academy of Sciences, CXVI, Art. 2 (July, 1964), 457-476.
This paper incorporates an illustrated outline of the stereochemical theo •
ry of olfaction, together with some historical information. Independent
lines of research which have confirmed this theory are reviewed.
Berglund, B., Berglund, V., Engen. T., and Lindvall, T. "The Effect of Adap-
tation on Odor Detection, " Perception and Psychophysics, IX (May, 1971),
435-438.
The sensitivity of human subjects to a weak odor following prior exposure
to it varying in both duration and concentration was studied. The effect
of adaptation was demonstrated, and the findings suggest that the sense
of smell is more stable than is usually believed.
Cain, William S. "Odor Intensity. Differences in the Exponent of the Psycho-
physical Function," Perception and Psychophysics^ VI (December, 1969),
349-354.
The results of a series of experiments showing that there are reliable
differences among the exponents of the psychophysical power functions
for odorants are discussed.
-212-
-------
Chzhen-tsi, Chzahao. "Basic Experimental Information for the Determination
of the Limit of Allowable Methanol Concentration in Atmospheric Air,"
in Limits of Allowable Concentrations of Atmospheric Pollutants, V. A.
Ryzaznov(ed.), Book 5 (Washington, D.C.: U. S. Public Health
Service, March, 1962), 67-68.
Studies were conducted of the presence and effects on rats and humans
of methanol as an atmospheric pollutant. An odor perception threshold
was determined for sensitive persons, and it was concluded that methanol-
emitting pharmaceutical and chemical plants should not be located near
residential areas and should be equipped with adequate control equipment.
Colueci,. Joseph M., and Barnes, Gerald J. "Evaluation of Vehicle Exhaust Gas
Odor Intensity Using Natural Dilution." Preprint, Society of Automotive
Engineers, Inc., New York, 1970.
A discussion of panel test procedures developed for field evaluation of
full-scale vehicle exhaust odor as well as odor tests in a controlled en-
vironment is presented. The procedure is capable of discriminating
between two odor sources of differing intensity.
Dravnieks, Andrew. "Physicochemical Basis of Olfaction," Annals of the New
York Academy of Sciences, CXVI, Art. 2 (July, 1964), 429-439.
A qualitative model of an olfactory chemosensor is proposed.
Ekman, Gosta, Berglund, Birgitta, Berglund, Ulf, and Lindvall, Thomas.
"Perceived Intensity of Odor as a Function of Time of Adaptation,"
Scandanavian Journal of Psychology, VII (1967), 177-186.
Two experiments were conducted to determine the perceived intensity
of odor under conditions of constant stimulation with hydrogen sulfide.
An exploratory investigation was also made of the process of recovery
of sensitivity after stimulation had ceased.
Engen, Trygg. "Effect of Practice and Instruction on Olfactory Thresholds,"
Perceptual and Motor Skills, X (June, 1960), 195.
This study concerned the measurement of olfactory thresholds by a
forced-choice method of limits. It was found that practice lowered
thresholds appreciably as did changes in the criterion of discrimination,
-213-
-------
. "Psychophysical Scaling of Odor Intensity and Quality," Annals of the
~New York Academy of Sciences, CXVI, Art. 2 (July, 1964), 504-516.
This paper summarizes some experiments with the direct psyehophysical
scaling methods.
_, and Pfaffmann, Carl. "Absolute Judgments of Odor Intensity," Journal
of Experimental Psychology, LVIII (July, 1959), 23-26.
This article discusses the ability of the human subject to transmit infor-
mation about odor intensity. The experiments concerned the greatest
number intensities of an odor that can be identified correctly by rank
order, and how the accuracy of identification is influenced by other fac-
tors, such as selection of odor ants, intensity of stimuli, practice, etc.
, and Pfaffmann, Carl. "Absolute Judgments of Odor Quality," Journal
of Experimental Psychology, LIX (April, 1960), 214-219.
This study was concerned with the recognition of odor qualities and the
application of information theory to the evaluation of the human observer,
Epstein, Samuel S. "Carcinogenicity of Organic Extracts of Atmospheric Pollu-
tants ,"JojJirjial_£fJ:hje^^ XVII, No. 11
(November, 1967), 728-729.
Results of a study of the carcinogenicity of atmospheric pollutants by the
injection of trace quantities of organic extracts in newborn mice are re-
ported . Experiments indicated a high potential hazard of carcinogenic
polluted air to man.
Foster, Dean. "Odor Correlation and the Panel Technique — 1953," Annals of
the New York Academy of Sciences, CXVI, Art. 2 (July, 1964), 482-483.
A brief review of the panel technique in 1953 and some of the progress
made between 1953 and 1963 are noted.
-214-
-------
Friberg, Lars. "Summary of Technical Problems." Paper presented at the
working conference on The Dose-Response Relationships Affecting Human
Reactions to Odorous Compounds, a symposium sponsored by the Environ-
mental Protection Agency, Air Pollution Control Office, Cambridge,
Massachusetts, April 26-27, 1971.
This paper discusses odor pollution in terms of the response, the dose,
and the dose-response relationships. Possible ways to approach the
various odor problems and the issues themselves are the focal points.
Friedman, Herman H., Mackay, Donald A., and Rosano, Henri D*. "Odor
Measurement Possibilities via Energy Changes in Cephalin Monolayers, "
Annals of the New York Academy of Sciences, CXVI, Art. 2 (July, 1964),
602-607.
After finding that the influence of air, in addition to the adsorption pro-
cess, during olfaction is important, the authors attempted to build an
artificial nose.
Gesteland, Robert C . "Initial Events of the Electro -Olfactogram, " Annals of
the New York Academy of Sciences, CXVI, Art. 2 (July, 1964), 440-447.
This paper presents the results of experiments using a wide variety of
stimulating substances. Conclusions about the early events associated
with EOG are also provided.
Hahn, John F. "Some Relations Between Smell and Other Sensory Response
'Systems, " Annals: of the New York Academy of Sciences, CXVI, Art 2
(July, 1964), 559-563.
Olfaction is discussed and several comparisons are made between the
phenomena associated with the other senses and that understood in the
sense of smell.
Hainer, Raymond M. "Some Suggested Critical Experiments in Olfactory
Theory, " Annals of the New York Academy of Sciences, CXVI, Art. 2
(July, 1964), 477-487.
This paper suggests several critical experiments to stimulate interest
in thinking about olfaction as a total perceptual system and in thinking
about extended applications in product research.
-215-
-------
Herbert, M., Click, R., and Black, H. "Olfactory Precipitants of Bronchial
Asthma," Journal of Psychosomatic Research, XI (August, 1967),
195-202.
The findings of this study indicate that olfactory stimuli play an impor-
tant role in precipitating asthmatic attacks .
Izmerov, N. F. "Atmospheric Air Pollution With Gasoline Vapor and Its Limit
of Allowable Concentration," in Limits of Allowable Concentrations of
Atmospheric Pollutants, V. A, Ryzaznov (ed.), Book 5 (Washington,
D.C.: U. S. Public Health Service, March, 1962), 52-67.
A report of a study concerning the effects of solvent gasoline vapors
emitted from crude oil on conditioned reflex reactions, odor thresholds,
and eye sensitivity is presented. Recommendations for allowable con-
centrations of low-sulfur gasoline vapor are presented as well as pro-
posed sanitary clearance zones around plants emitting such vapors.
Jones, F. Norvell. "Subjective Scales of Intensity for Three Odors, " American
Journal of Psychology, LXXI(June, 1958), 423-425.
The subjective intensities of different concentrations of benzene, heptane,
and octane were scaled by the method of magnitude estimation, This
study showed that olfactory intensity obeys the law proposed by Stevens
for subjective intensities.
Karneyev, Yu E . "Determination of Allowable Maximum Concentrations of Phenol
and Acetophenone When Simultaneously Present in Atmospheric Air," in
The Biological Effects and Hygienic Importance of Atmospheric Pollutants,
Book 10 (U.S.S.R. Literature on Air Pollution and Related Occupational
Diseases, 1968), XVII, 97-107.
The combined effect of phenol and acetophenone vapors on the human
organism was studied. Odor threshold levels for sensitive persons are
presented.
-216-
-------
Kendall, David A., and Jacobs, Stanley P. "Physiological Responses to Odor -
ants ." Paper presented at the working conference on the Dose-Response
Relationships Affecting Human Reactions to Odorour Compounds, a sym -
posium sponsored by the Environmental Protection Agency, Air Pollution
, Control Office, Cambridge, Massachusetts, April 26-27, 1971.
This paper reports the results of a study where odorants were presented
to subjects on a totally blind basis, while concentrations and duration of
stimulus were controlled and a number of autonomic responses were
monitored. .
Mayers, Martin A. "Odor Technology: Where We Are, Where We Are Going,"
Annals of the New York Academy of Sciences, CXVI, Art. 2 (July, 1964),
369-377. :
A review of the knowledge of the olfactory system, measurement and
;control, and some recommendations for further study are provided.
McNamara, B. P., and Danker, W. M. "Odors and Taste, " in Basic Principles
• c of Sensory Evaluation (American Society for Testing and Materials,
Philadelphia, Pennsylvania, STP-433, 1968),.13-18.
Difficulties encountered in evaluating response to odor and taste are
discussed. Theories of the olfactory sense are discussed.
Mitchell, M. J., and Gregson, R. A. "Stimulus Specificing and Thresholds of
. the Chemical Senses ." Research Project, University of Canterbury,
New Zealand (1968), No > 14. .
This paper is a report of a study where subjects took 10 ml. of ethanol
orally. Results indicate that marked changes in psychophysical behavior
occur when small amounts of some contaminants are present in absolute
threshold determinations . Impurities are found to influence olfactory
thresholds. .••• , • . • ..• '<
Moulton, David G. "Nasal Responses to Chemical Stimulation." Final Report.
Pennsylvania University, Philadelphia, Mone 11 Chemical Senses Center,
Grant DAHC 19-68-G-0026, Report 1, September, 1969. :-
This paper discusses a study on the properties of the nasal trigeminal
and olfactory systems which were explored in rat and man using electro-
physiological, anatomical, and behavioral techniques .
-217-
-------
, and Tucker, D. "Electrophysiology of the Olfactory System,"
Annals of the New York Academy of Sciences, CXVI, Art, 2 (July,
1964), 380-428.
A survey of studies on the electrical activity associated with stimula-
tion of the odor receptors and accessory structures is presented. An
extensive bibliography is provided.
Ottoson, David G, R „ "How We Recognize Odors," New Scientist, XLVIII.
No. 723(October, 1970), 114-116, T— -
The direct correlation between the subjective sensation of small and
the electrical activity in particular cells of the nasal cavity is discussed.
Findings from recent research are indicated.
Rosano, Henri L., and Scheps, Sheldon Q. "Adsorption-Induced JBlectrode
Potential in Relation to Olfaction," Annals of the New York Academy of
Sciences, CXVI, Art. 2 (July, 1964), 590-601.
The authors attempted to develop an artificial nose and found that gal-
vanic phenomena can be applied to gas and vapor analysis. The working
conditions and a probable mechanism, to be used are advanced.
Ryozanov, V. A. "Sensory Physiology as Basis for Air Quality Standards, "
Archives of Environmental Health, V (November, 1962), 480-494.
A discussion of the need for clean air, a method for determining some
odor thresholds, effects of odors on the respiratory system, and re-
sults of several studies regarding the influence of aromatic substances
on the human central nervous system are presented.
Schneider, Robert A., and Wolf, Stewart. "Olfactory Perception Thresholds
for Citral Utilizing a New Type Olfactorium," Journal of Applied
Physiology, VIII (1955), 337-342.
This paper describes a relatively odor'free, temperature and humidity
controlled, walk-in olfactorium and preliminary observations on odor
perception thresholds using this olfactorium.
-218'
-------
Schutz, Howard G. "Odor Theory, Physiology, and Chemistry, " Annals.of the
New York Academy of Sciences, CXVI, Art. 2 (July, 1964), 378-379.
Ideas that were indicated in the 1963 Odor Conference Held at the
New York Academy of Sciences are briefly noted.
Springer, Karl J., and Dietzmann, Harry E . "Correlation Studies of Diesel
Exhaust Odor Measured by Instrumental Methods to Human Odor Panel
Ratings: Progress Report.".. Paper presented at the Conference on Methods
for Measuring and Evaluating Odorous Air Pollutants at the Source and in
the Ambient Air, Stockholm, June 1-5, 1970.
This discussion centers around attempts to extend the usefulness of the
trained human odor panel by means of a chemical-instrumental method
developed at the Southwest Research Institute, San Antonio, Texas.
Stone, Herbert, and Dryer, Gordon. "Some Properties of the Olfactory System
of Man, " Perception and Psychophysics, II (November, 1967), 516-518.
Preliminary investigations into the effects of a twelve-hour diurnal vari-
ation suggested there was a tenfold decrease in odor sensitivity from
morning to evening. Blindfolding subjects and auditing stress on the
'test-situation did not. significantly alter odor sensitivity.
Terraglio, Frank P., and Manganelli, Raymond M.. .''The Absorption.of Atmos-
pheric Sulfur Dioxide by Water Solutions," Journal of the Air Pollution
Control Association, XVII, No. 6 (June, 1967), 403-406.
This article presents a technical discussion of a laboratory study to
determine the rate and extent of absorption of atmospheric sulfur diox-
ide by distilled water surfaces . The information was used in part to
determine the possible health effects concerning solutions of acid gases
-219-
-------
Tkach, N, Z . "Combined Effects of Low Acetone and Acetophenone Concentra-
tions In the Air of the Living Organism," in The Biological Effects and
Hygienic Importance of Atmospheric Pollutants, Book 10 (U.S.S.R,
Literature on Air Pollution and Related Occupational Diseases, 1968),
XVII, 108-117,
The effects of low acetone and acetophenone concentrations on the living
organism were determined. The odor perception threshold concentration
was determined using the most sensitive test persons for acetone, for
acetophenone, and for a mixture of these two compounds •„ Recommended
minimum perceptible concentrations are presented,
Wilby, F. V, "Variation in Recognition Odor Threshold of a Panel," Journal of
the Air Pollution Control Association, XIX (February, 1969), 96-4QO,
This paper is a report of a study designed to correlate the odor strength
of natural gas with its sulfur analysis. The recognitipn odor thresholds
of 18 sulfur compounds were determined,, using an untrained panel of
35 people.
Wright, Robert Hamilton. "Odor and Molecular Vibration: The Far Infrared
Spectra of Some Perfume Chemicals, " Annals of the New York Academy
of Sciences, CXV1, Art. 2 (July, 1964), 552-558. ~~~~
The vibrational theory of olfaction is discussed in detail. Results of
studies lend some support t
-------
C. ODOR MEASUREMENT AND EVALUATION TECHNIQUES
Adams, D. F., Koppe, R. K., and Tuttle, W. N. "Analysis of Kraft-Mill Sulfur
Containing Gases With GLC lonization Detectors, " Journal of the Air
Pollution Control Association, XV, No. 1 (January, 1965), 31-33.
This paper describes a technique which includes the use of two chroma-
tographic columns in series to separate O2, N2, CO, H2O, H2S, and
CHsSH.i, The conditions required to obtain adequate sensitivity and
separations are discussed.
"Physical-Chemical Measurements of Odorous Pollutants, " Paper
presented at the Conference on Methods for Measuring and Evaluating
Odorous Air Pollutants at the Source and in the Ambient Air, Stockholm,
June 1-5, 1970.
A discussion of the analytical methodologies available for measuring
those sulfur containing gases which are most significant to the kraft
pulp industry is presented. The article traces instrumental analysis
in two major areas: source gases and ambient air.
Baker, Robert A. "Response Parameters Including Synergism-Antagonism in
Aqueous Odor Measurement," Annals of the New York Academy of
Sciences, CXVI, Art. 2 (July, 1964), 495-503.
This paper reviews a program of study concerning odor in water, the
existing test methods, and the effect of variables, interactions, and
complex mixtures.
Beck, Lloyd Henry. "A Quantitative Theory of the Olfactory Threshold Based
Upon the Amount of the Sense Cell Covered by an Adsorbed Film,"
Annals of the New York Academy of Sciences, CXVI, Art. 2 (July, 1964),
448-456.
The author discusses the development of a hypothesis concerning a rela-
tionship between molecular structures and the threshold concentration
of odorous substances so that the threshold is predictable.
-221-
-------
Bemforado, D. M., Rotella, W. J., and Haton, D. L. "Development of an Odor
Panel for Evaluation of Odor Control Equipment," Journal of the Air Pollu-
tion Control Association, XIX (February, 1969), 101-105.
Test data show that an odor evaluation panel using the syringe dilution
technique is a useful and practical tool in quantitatively analyzing indus-
trial odors.
Berglund, B., Berglund, V., Ekman, G., and Engen, T. "Individual Psycho -
physical Functions for Twenty-Eight Odorants," Perception and Psycho-
physics, IX (March, 1971), 379-384.
Individual scales of odor intensity were obtained for 28 different chemi-
cal compounds by the method of magnitude estimation. Additional
results of the experiments for this study, concerning power functions
which describe the relationship between partial vapor pressure of the
odorants and their subjective odor intensity, are also provided.
Bozza, G., Calearo, C., and Teatini, G. P. "On the Making of a Rational
Olfactometer, " Acta Qto-Laryngologica, LII (September, 1960),
189-209.
A new olfactometer is extensively described. Some of the main
features are: (1) exact and constant concentration of odorous sub-
stances, (2) stimulation in natural conditions, and (3) the possibility
of changing stimulus during the interval between one inspiration and
the following one.
Brosset, Cyrill. "The Stability of Emitted Odorous Substances in the Atmos-
phere." Paper presented at the Conference on Methods for Measuring
and Evaluating Odorous Air Pollutants at the Source and in the Ambient
Air, Stockholm, June 1-5, 1970.
A summary of the reactions on the atmospheric chemistry of emitted
odorous substances is provided. The compounds discussed in this article
are those of significance in the emission from sulfate mills.
Byrd, J. F. "Demonstration — Syringe Odor Measurement Technique," Journal
of the Air Pollution Control Association, LVIII (May, 1957), 58-5
-------
Cain, William S. "Odorant Differences in the Scaling of Olfactory Intensity, "
Psychonomic Bulletin , 1(1967), 8.
Direct scaling od odor intensity with an air-dilution olfactometer extend-
ed previous results that the water solubility of odorants is correlated
with steepness of intensity scales .
Cerderlof, Rune, Edfors, Marie-Louise, Friberg, Lars, and Lindvall, Thomas,
"On the Determination of Odor Thresholds in Air Pollution Control — An
Experimental Field Study on Flue Gases From Sulfate Cellulose Plants,"
Journal of the Air Pollution Control Association, XVI (February, 1966),
92-94.
A field method designed for estimating the odor thresholds of flue gases
emitted from different industrial processes is discussed. Results of two
studies and a discussion of the practical application of the method are
provided.
Czerwonka, L. J., Boylan, R. J., and Gonzalez, D. J. "Accelerated Test
Method for Determination of Odor Control Devices, " Annals of the New
York Academy of Sciences, CXVI, Art 2 (July, 1964), 613-620.
Gas chromatographic devices were investigated as an accelerated test
method for evaluating sorptive beds. This paper reviews one accelerated
test method under study.
Doving, KjellB., and Lange, Anders L. "Comparative Studies of Sensory
Relatedness of Odors, " Scandanavian Journal of Psychology, VIII,
(1967), 47-51.
Measures of relatedness of odors were obtained by means of two psycho-
physical methods.and one electrophysiological technique . Significant
rank correlations were obtained between the results of the three methods,
Dravnieks, Andrew, and Laffort, Paul. "Possibilities of Predicting Odor Inten-
sity and Quality by Means of Chemical Analysis ." Paper presented at
the Conference on Methods for Measuring and Evaluating Odorous Air
Pollutants at the Source and in the Ambient Air, Stockholm, June 1-5, 1970,
A discussion of the suitable techniques and correlations which have
emerged concerning the translation of chemical analytical data into
organoleptic parameters describing odor intensity and quality is pre-
sented .
-223-
-------
Duffee, R. A. "Appraisal of Odor-Measurement Techniques, " Journal of the
Air Pollution Control Association, XVIII (July, 1968), 472-474.
Relative advantages and disadvantages of odor measurement methods
are discussed. Applications and suggested modifications are also
presented.
Ekman, Gosta. "A Direct Method for Multidimensional Ratio Scaling," Psycho-
metrika, XXVIII (March, 1963), 33-41.
A generalization of direct ratio scaling methods for multidimensional
ratio scaling is described. The method was developed to meet require-
ments for experimentation in several areas including olfaction.
, Berglund, Birgitta, Berglund, Ulf, and Lindvall, Thomas. "Perceived
Intensity of Odor as a Function of Time of Adaptations, " Scandanavian
Journal of Psychology , VIII (1967), 177-186.
The perceived intensity of odor under conditions of constant stimulation
with hydrogen sulfide was investigated in two experiments. Perceptional
intensity was found to decrease exponentially with increase of time and
stimulation.
Engen, Trygg. "Organoleptic Measurements of Odor Qualities ." Paper pre-
sented at the Conference on Methods for Measuring and Evaluating
Odorous Air Pollutants at the Source and in the Ambient Air, Stockholm,
June 1-5, 1970.
A discussion of the recent psychophysical approach to the quality of
odors which rejects verbal labels and attempts instead to increase the
generality of the data by simplifying the responses and testing them
against multidimensional mathemetical models is presented.
and McBurney, Donald H. "Magnitude and Category Scales of the
Pleasantness of Odors, " Journal of Experimental Psychology, LXVIII
(November, 1964),. 435-440.
The pleasantness of odors was judged by the method of magnitude and
the category method. Comparison of the results of the two methods
suggests a reason for the general lack of agreement between them.
-224-
-------
Foster, D., Scofield, E. H., and Dallenbach, K. "An Olfactorium," American
Journal of Psychology, LXIII (July, 1950), 431.
A detailed description and explanation of an olfactorium, an odor-proof
cubical, is presented. This cubical serves as a laboratory for qualita-
tive research and as part of an olfactometer for quantitative investiga-
tions .
Fox, E. A., and Gex, V. E. "Procedure for Measuring Odor Concentration in
Air and Gases," Journal of the Air Pollution Control Association, LVIII
(May, 1957), 60-61.
A procedure for determining the odor concentration in the atmosphere
and in gases discharged from industrial process operations and for
determining the odor emission rate from a stack or a vent is presented
in detail.
Gerstein, H. H. "A Continuous Odor Monitor and Threshold Tester," American
Water Works Association Journal, XLIII (May, 1951), 373-377.
Use of a continuous odor monitor and threshold tester which provides a
quick means for determining the type and intensity of an odor at any
given time and also makes possible a frequent check on the efficiency
of the treatment process is presented.
Hemeori, Wesley C. L. "Technique and Apparatus for Quantitative Measurement
of Odor Emissions," Journal of the Air Pollution Control Association,
XVIII (March, 1968), 166-170.
A description of the instruments and techniques developed by the author
over a period of years for the quantitative measurement of odor emis-
sions at the source is presented. The system is based on quantitative
dilution and human subjects . It is a dynamic method which emphasizes
a large volume of samples and an odormeter to avoid loss of odor on the
walls.
Hendrickson, E. R. "Needs for Future Research." Paper presented at the
Conference on Methods for Measuring and Evaluating Odorous Air Pollu-
tants at the Source and in the Ambient Air, Stockholm, June 1-5, 1970.
This paper outlines several crucial research problems concerning the
measurement and evaluation of odors which must be solved before
odorous emissions can be controlled.
-225-
-------
Henion, Karl E. "Psychophysical Scales of the Olfactory Pleasantness of Homo-
logous Alcohols," P£r£e^tiwi_jnd^Ps^chophysic£, IX (February, 1971),
234-236.
The results of a study which established a category, a ratio estimation,
and three magnitude scales from the observations of subjects who judged
the pleasantness of a homologous series of naliphatic alcohols is pre-
sented .
. "Subjective Range of the Odor Oiliness of Heptanol," Journal of
"Experimental Psychology. LXXXII (December, 1969), 515-519.
The results of experiments on the oiliness of agrometric dilution series
of N-heptanol are presented as well as a category, a ratio estimation,
and three magnitude scales .
Hogstrom, Ulf. "Predicting Odor Frequencies by Dispersion Calculations ."
Paper presented at the working conference on The Dose-Response Rela-
tionships Affecting Human Reactions to Odorous Compounds, a sympo-
sium sponsored by the Environmental Protection Agency, Air Pollution
Control Office, Cambridge, Massachusetts, April 26-27, 1971.
The problem of making odor dispersion calculations is discussed and a
method for making such calculations is advanced.
Horstman, S. W., Wromble, R. F., and Heller, A. N. "Identification of
Community Odor Problems By Use of an Observer Corps," Journal of
the Air Pollution Control Association, XV, No. 7 (June, 1965), 261-264.
Approximately 120 high school students were used to collect odor data.
Data were reviewed to show severely affected areas, objectionable odor
types, frequency of odors, and associated weather parameters .
Huey, N., Broering, L., Jutze, G., and Gruber, C. "Objective Odor Pollution
Control Investigations," Journal of the Air Pollution Control Association,
X (December, I960), 441.
Objective odor pollution control investigations associated with various
manufacturing processes are the topic of this article. Included in the
article are discussions of: The Nature of Ambient Odors, Measurement
of Odor Strength — the Scentometer, Source Location, Standard Proce-
dure for Ambient Odor Investigation, and Meteorological Effects on Odor
Nuisance Occurrences.
-226-
-------
Hum, R. W. "Source and Measurement of Odorous Engine Products." Paper
presented at the Conference on Methods for Measuring and Evaluating
Odorous Air Pollutants at the Source and in the Ambient Air, Stockholm,
June 1-5, 1970.
This paper is a review of studies concerning engine product odors. The
author presents this review within the context of all combustion of hydro-
carbon materials.
Jacobs, Morris B. "Recommended Standard Method for Continuing Air Monitor-
ing for Hydrogen Sulfide Ultramicrodetermination of Sulfides in the Air, "
Journal of the Air Pollution Control Association, XV, No. 7 (July, 1965),
314-315.
A discussion of the methylene blue method for estimating the concentra-
tion of trapped sulfides is presented. A colorimetric method and a
spectrophotometric method are described.
Johnson, Kenneth D. "Odorant Sources in Air Pollution." Paper presented at
the working conference on The Dose-Response Relationships Affecting
Human Reactions to Odorous Compounds, a symposium sponsored by the
Environmental Protection Agency, Air Pollution Control Office,
Cambridge, Massachusetts, April 26-27, 1971.
This paper provides a general overview of odorant sources .
Johnston, James W., Jr. "Organoleptic Analyses of Putrid Olfactory Stimuli, "
Psychonomatic Bulletin, 1(1967), 9.
Significance of results obtained when several putrid smelling compounds
are reacted in oxidents and heavy metal sets in aqueous solution in order
to destroy or alter the quality are discussed.
and Rubacky, Eugene P. "Sensory Measurements of a Hedonic Tone
and Odor Quality." Paper presented at the Conference on Methods for
Measuring and Evaluating Odorous Air Pollutants at the Source and in
the Ambient Air, Stockholm, June 1-5, 1970.
The authors explain a scalar technique employing a panel of 10 to 15
judges to acquire specific, technical data for the evaluation of atmos-
pheric and source odors.
-227-
-------
Jones, F. Norvell. "A Comparison of the Methods of Olfactory Stimulation:
Blasting vs. Sniffing, " American Journal of Psychology, LXVIII
(September, 1955), 486.
A discussion of the controlled blast method and sniffing for the collection
of threshold data is presented. It is suggested that additional research
is called for before both methods can be considered to have the same
meaning.
. "Scales of Subjective Intensity for Odors of Diverse Chemical
"Nature, " American Journal of Psychology. LXXI (March, 1958), 305-310,
The subjective intensities of N-butanol, sec-butanol, iso-butanol, ethyl
acetate, cyclohexane, and pyridine were scaled by the method of magni-
tude estimation. The resultsof this study showed that there is no
evidence for a different stimulus-receptor interaction for polar as
compared to non-polar substances.
. "The Forced-Choice Method of Limits, " American Journal of Psy-
"chology, LXIX (December, 1956), 672.
The forced-choice method of limits, a modification of the indirectional
method of limits technique of determining olfactory thresholds, is dis-
cussed.
. "The Reliability of Olfactory Thresholds Obtained By Sniffing, "
American Journal of Psychology, LXVIII (June, 1955), 289.
A discussion of a study designed to determine if sniffing is a reliable
method for determining thresholds of different odorants is provided.
This study determined that sniffing is an adequate technique.
, and Woskow, Morris H. "On the Intensity of Odor Mixtures," Annals
of the New York Academy of Sciences. CXVI, Art. 2 (July, 1964),
484-494.
This article discusses experiments and results of a study designed to
determine the interactions between pairs of odors above the threshold
levels.
-228-
-------
Kendall, David A., and Neilson, Anne J. "Correlation of Subjective and Objec-
tive Odor Responses," Annals of the New York Academy of Sciences,
CXVI, Art. 2 (July, 1964), 567-575.
Results of tests on a range of concentrations of four odorants in two
media with an expert odor panel and gas-liquid chromatography are
presented. Comparative data on odor threshold concentrations,
changes in odor characteristics, and intensity produced by normal
contaminants present in the odorants and by known mixtures of the
odorants are also provided.
Leonardos, G., Kendall, D,, and Barnard, N. "Odor Thresholds Determina-
tions of 53 Odorant Chemicals," Journal of the Air Pollution Control
Association, XIX (February, 1969), 91-95.
In order, to assist in assessing potential odor problems arising from
chemical manufacturing operations, the odor thresholds of 53 commer-
cially important odorant chemicals were determined using a standard-
ized and defined procedure.
Lindvall, Thomas . "Sensory Evaluation of Odor Intensity at the Source and in
the Ambient Air." Paper presented at the Conference on Methods for
Measuring and Evaluating Odorous Air Pollutants at the Source and in
the Ambient Air, Stockholm, June 1-5, 1970.
This paper discusses the aims of environmental hygienic evaluations of
odorous air pollutants, requirements for the procedure used, and equip-
ment and methods for sensory source strength analysis .
.. "Sensory Evaluation of Odors in the Ambient Air." Paper presented
at the working conference on The Dose-Response Relationships Affecting
Human Reactions to Odorous Compounds, a symposium sponsored by the
Environmental Protection Agency, Air Pollution Control Office,
Cambridge, Massachusetts, April 26-27, 1971.
Sensory methodology for ambient air analysis is discussed, including
the following topics: equipment, sensory methods, and sensory evalu-
ation in the ambient air.
-229-
-------
Lyshkow, Norman A. "The Continuous Analysis of Sulfur Dioxide in Gaseous
Samples," Journal of the Air Pollution Control Association, XVII, No. 10
(October, 1967), 687-689.
A technical discussion of the achievement of the stability required for
the continuous analysis of sulfur dioxide by a reagent is presented.
Mateson, J. F. "Olfactometry: Its Technique and Apparatus," Journal of the
Air Pollution Control Association, VI (November, 1955), 167-170.
A short review and classification of odor measurement devices is pre-
sented. New devices being developed are described, and factors
effecting organoleptic odor determinations are outlined.
McGinnity, J. L., Grimley, K. W., Jr., Horres, C. R., Jr., and Mulik, J. D,
"Mobile Source Sampling Laboratory for the Pulp and Paper Industry."
Paper presented at the Conference on Methods for Measuring and
Evaluating Odorous Air Pollutants at the Source and in the Ambient
Air, Stockholm, June 1-5, 1970.
This paper discusses a mobile source -sampling laboratory designed to
be employed in the source testing program of a study for the pulp and
paper industries in the United States .
"Methods for Measuring and Evaluating Odorous Air Pollutants at the Source and
in the Ambient Air." Report of an International Symposium in Stockholm,
June 1-5, 1970.
This report is a result of discussions during a recent technical sympo-
sium. The report represents the consensus of the opinions of the
participants and covers the evaluation of odorous air pollutants, effects
of such pollutants, and recommendations for future research.
Mills, John L, Walsh, Robert T., Luedtke, KarlD., and Smith, Lewis K.
"Quantitative Odor Measurement, " Journal of the Air Pollution Control
Association, XIII (October, 1963), 467-475. " ' '
Modifications to the ASTM odor measurement method are described.
Methods for controlling odors in certain industries are discussed.
-230-
-------
Montgomery, T. L., and Corn, M. "Adherence of Sulfur Dioxide Concentra-
tions in tKe Vicinity of a Steam Plant to Plume Dispersion Models,"
Journal of the Air Pollution Control Association, XVII, No. 8 (August,
1967), 512-517.
A technical explanation of the present dispersion models concerning
their varibility in predicting concentrations of SC-2 is presented. The
study suggests that present dispersion models should be used only when
predicting maximum concentrations under specific meteorological con-
ditions .
Munn, R. E ., and Stewart, I. M. "The Use of Meteorological Towers in Urban
Air Pollution Programs," Journal of the Air Pollution Control Association,
XVII, No. 2 (February, 1967), 98-101.
This paper provides a discussion of the relevance and application of data
from the use of meteorological towers in the study of air pollution.
Nickol, G. B. "Rank Odor Method for Evaluating Stack Gases." Journal of the
Air Pollution Control Association, LVIII (May, 1957), 55.
A simple, reliable procedure for evaluating stack gases is discussed.
Nolan, Philip. "Present-Day Limits on Infrared Absorption Techniques in
Measurement of Odor, " Annals of the New York Academy of Sciences,
CXVI, Art. 2 (July, 1964), 576-582.
This paper describes the limits of performance in determining trace
concentrations of organic impurities in both large and small volumes
of normal air.
Panghorn, R. M., Berg, H. W., Roessler, E. B., and Welsh, A. D. "Influence
of Methodology on Olfactory Response," Perceptual and Motor Skills,
XVHI (February, 1964), 91-103.
The odor of 2-heptanone was delivered by a precision olfactometer to
subjects using five methods of single-sample presentation of concentra-
tions. These methods arid specific modifications of the constant-
stimulus method are discussed.
-231-
-------
Reese, T. S., and Stevens, S. S. "Subjective Intensity of Coffee Odor, "
American Journal of Psychology, LXXIII (September, 1960), 424-428.
The aim of this study was to develop a method of presenting controlled
olfactory stimuli derived from natural coffee. It also was designed to
attempt to determine how the subjective intensity of the odor grows with
stimulus -concentration.
Rosen, A. A., Peter, J. B., and Middleton, F. M. "Odor Thresholds of Mixed
Organic Chemicals, " Journal of the Water Pollution Control Federation.
XXXIV (January, 1962), 7-14.
Results of a series of experiments on the odor thresholds of some chemi-
cals and a discussion of odor synergism are presented.
, Skeel, R. T., and Ettinger, M. B. "Relationship of River Water Odor
to Specific Organic Contaminants, " Journal of the Water Pollution Control
Federation, XXXV (June, 1963), 777-782.
This report describes the results of odor tests conducted in conjunction
with a study of organic contaminants . The phenomenon of odor syner-
gism was demonstrated in the polluted river that was tested.
Rumsey, Donald W., and Cesta, Ramon P. "Odor Threshold Levels for UDMH
andNO2-" American Industrial Hygiene Association Journal, XXXI,
No. 3 (May-June, 1970), 339-343.
This report discusses seven years of field experience that indicated
that the actual threshold levels for unsymmetrical dimethylhydrazine
(UMDH) and nitrogen dioxide are considerably lower than those gener-
ally quoted.
Schiff, William, and Thayer, Stephen. "Cognitive and Affective Factors in
Temporal Experience: Anticipated or Experienced Pleasant and Un-
pleasant Sensory Events." Perceptual and Motor Skills, XXVI (May,
1968), 799-808.
A discussion of a study where subjects reproduced and verbally esti-
mated short or long durations under conditions of smelling or antici-
pating smelling a pleasant, unpleasant, or neutral substance.
-232-
-------
Schutz, Howard G. "A Matching-Standards Method for Characterizing Odor
Qualities, " Annals of the New York Academy of Sciences, CXVI, Art. 2
(July, 1964), 517-526.
The results of a study showing that a matching-standards technique for
describing the qualitative aspects of an odor is a reliable and valid
method are presented.
Skinner, C. F. "Odor Control," National Engineer, LXXIV, No. 7 (July, 1970),
12-14.
A dilution method using the human nose to measure odor concentration
and catalytic and thermal incineration systems are discussed.
Stone, Herbert. "Behavioral Aspects of Absolute arid Differential Olfactory Sen-
sitivity, " Aiu]jdsj:)fj:he_J^e^ CXVI, Art. 2
(July, 1964), 527-534.
Results of studies conducted to test Weber's law and to reevaluate
Fechner's stimulus-response relationship are presented. Evaluations
of the olfactometer are also presented.
__, Ough, C. S., and Panghorn, R. M. "Determination of Odor Difference
Thresholds, " Journal of Food and Science, XXVII (1962), 197-202.
A modified olfactometer was used to determine odor difference thresholds
at four levels of 2 -heptanone. The apparatus was found to give rapid and
reproducible threshold measurements.
Straforelll, J. B., Paszner, L., and Wilson, J. W. "Radiolysis of Kraft Mill
Gaseous Effluents." Preprint, Canadian Pulp and Paper Association,
Technical Section, and Chemical Institute of Canada (1970), 57.
Gas chromatographic studies of radiolytic effects of atomic radiation
from a cobalt source on model compounds of kraft mill gaseous efflu-
ents are noted. An hypothesis for the process mechanism is suggested.
-233-
-------
Stuiver, M. "An Olfactometer With a Wide Range of Possibilities, " Acta Oto-
Laryngologica, LI (January-February, 1960), 135-142.
An olfactorium which is very suitable for threshold measurements of
the organ of smell is described. Rate of flow of odorous air, duration
of stimulus, and number of odorous molecules are controlled inde-
pendly.
Sullivan, D. C., Adams, D. F., and Young, F. A. "Design of an 'Odor Percep-
tion and Objectionability Threshold' Test Facility, " Atmospheric Environ-
ment, II (March, 1968), 121-133.
The design requirements for the study of the thresholds of odor "percep-
tion" and "objectionability" in a large population of human subjects are
discussed. Collection of data was facilitated by converting a bus into an
air-conditioned, mobile odor perception laboratory.
Terabe, Moto J., Oomichi, Sadao, Benson, F. B., Newill, V. A., and Thompson,
J. E . "Relationships Between Sulfur Dioxide Concentration Determined By
the West-Gaeke and Electroconductivity Methods, " Journal of the Air
Pollution Control Association, XVII, No. 10 (October, 1967), 673-675.
Relationships between the sulfur dioxide concentrations determined by
the West-Gaeke and electroconductivity methods are discussed. SO2
concentrations determined by the electroconductivity method were
greater, particularly in the winter season. The tests were held in
Kawasaki, Japan, as a part of the United States-Japan Cooperative Air
Pollution Measurement Studies.
Teranishi, Roy, Buttery, RonG., and Mon, T. R. "Direct Vapor Analyses With
Gas Chromatography, " Annals of the New York Academy of Sciences,
CXVI, Art. 2 (July, 1964), 583-589.
Some of the problems encountered in detecting and recording food
volatiles, the parameters which must be considered in order to
utilize, the sensitivity possible with ionization detectprs, and some
examples of direct vapor sampling analyses are presented and dis-
cussed.
-234-
-------
Turk, Amos. "Approaches to Sensory Odor Measurement, " Annals of the New
York Academy of Sciences, CXVI, Art. 2 (July, 1964), 564-566.
A critical evaluation of the methods used to establish odor threshold
values is presented. The article covers extrapolation of dilutions
ratios, analysis of odorant solutions, tracer methods, and presenta-
tion of odor stimuli to subjects.
. "Odor Source Inventories." Symposium on Odor Control, Sixty-
Seventh National Meeting, Atlanta, Georgia, February 15-18, 1970.
Several methods used to compile odor source inventories are reviewed
and discussed in detail. A procedure for translating quality descrip-
tions into inventories of odor sources is presented.
. "Problems in Sampling and Storing Stack Gases for Chemical and
Sensory Analysis." Paper presented at the Conference on Methods for
Measuring and Evaluating Odorous Air Pollutants at the Source and in
the Ambient Air, Stockholm, June 1-5, 1970.
Since stack gases are usually too hot and/or concentrated for direct
presentation to people for sensory analysis, this paper considers the
general problem of diluting odorants for sensory analysis as well as
the various methods of sampling for chemical analysis.
. "Properties of Odorants ." Paper presented at the working confer-
ence on The Dose-Response Relationships Affecting Human Reactions
to Odorous Compounds, a symposium sponsored by the Environmental
Protection Agency, Air Pollution Control Office, Cambridge,
Massachusetts, April 26-27, 1971.
This paper discusses odor sources that can be classified according to
their degree of confinement, character or quality, hedonic impact, and
persistence or pervasiveness in the atmosphere.
-235-
-------
Ubaydullayev, R. "Atmospheric Air Pollution With Vapors of Hydrolytic Alco-
hol and Its Effect on the Organism, " in The Biological Effect and Hygienic
Importance of Atmospheric Pollutants, Book 10 (U.S.S.R. Literature on
Air Pollution and Related Occupational Diseases, 1968), XVII, 46-54.
This report presents the threshold of ethanol odor perception in most
sensitive persons. Recommended minimum perceptible concentrations
are presented.
Vogh, J. W. "Nature of Odor Components in Diesel Exhaust, " Journal of the
Air Pollution Control Association, XIX (October, 1969), 773-777.
The assessments of a human panel were used to determine odorant
identity and quantitative contribution to exhaust odor.
Waller, R. A. "Environmental Quality, Its Measurement and Control,"
Regional Studies, IV (August, 1970), 177-191.
This paper summarizes techniques of environmental quality evaluation
and recommends a method by which many diverse aspects of the
environment, including odors, can be related to a common scale which,
in turn, can have a monetary value attached to it.
Wilkens, Walter F ., and Hartman, John D. "An Electronic Analog for the Olfac-
tory Processes, " Annals of the New York Academy of Sciences, CXVI,
Art. 2 (July, 1964), 608-612.
A discussion of refinements to an electrochemical device developed as a
form of instrumentation for rapid and objective measurement of volatile
flavors and other odors is provided.
Wohlers, Henry C . "Odor Intensity and Odor Travel From Industrial Sources, "
International Journal of Air and Water Pollution, VII (May, 1963), 71-78.
Odor surveys were completed of the stack effluents in the vicinity of a
petroleum coking plant, a kraft paper mill, an onion and garlic dehydra-
tion plant, and a rotogravure printing plant. Extent of odor travel was
compared with the calculated threshold dilution of the stack gases accord-
ing to Sullivan's equation. Actual measurements did not always agree
with calculated dilutions.
-------
. "Recommended Procedures for Measuring Odorous Contaminants in
"the Field," Journal of the Air Pollution Control Association, XVII, No. 9
(September, 1967), 609-613.
A description of generalized facts concerning odors is presented. Odor-
ous measurement methods for individuals with a working knowledge of
either odor or air pollution problems are discussed.
Wright, Robert Huey, and Michels, Kenneth M. "Evaluation of Far Infrared
Relations to Odor by a Standards Similarity Method, " Annals of the New
York Academy of Sciences, CXVI, Art. 2 (July, 1964), 535-551.
Results of a rigorous experiment evaluation of the relationship between
odor and low-frequency molecular vibrations are presented. The method
and chemical selection for the study were guided by the information found
in a literature evaluation.
Yuldashev, T. "Toxicity of Ethylene Oxide in Low Concentrations," in The Bio-
logical Effects and Hygienic Importance of Atmospheric Pollutants, Book
10 (U.S.S.R. Literature on Air Pollution and Related Occupational
Diseases, 1968), XVII, 33-39.
This report presents the threshold of ethylene oxide odor perception in
most sensitive test persons. Recommended minimum perceptible con-
centrations are presented.
-237-
-------
D. ODOR ABATEMENT AND CONTROL TECHNOLOGY
Acres, G. J. K. "The Control of Air Pollution. Platinum Catalyst Systems for
' Industrial Odor Control, " Platinum Metals Review, XV, No. 1 (January,
1971), 9-12.
A honeycomb catalytic system with heat recovery is schematically illus-
trated and discussed. The system has been successfully evaluated for
the removal of odors from animal, fish, and food processing industries.
Barnebey, H. L. "Removal of Exhaust Odors From Solvent Extraction Operation
By Activated Charcoal Adsorption, " Journal of the Air Pollution Control
Association, XV, No. 9 (September, 1965), 422.
A discussion of the method of operation used in the removal of isopropyl
alcohol vapor from the air and the reuse of the recovered alcohol in the
solvent extraction process is presented.
Bloomfield, Bernard D. "Costs, Efficiencies, and Unsolved Problems of Air
Pollution Control Equipment, " Journal of the Air Pollution Control Assoc-
iation, XVII, No. 1 (January, 1967), 28-32.
A detailed discussion of the cost of air cleaning equipment operational
and performance problems is provided.
Bosworth, Cyrus M., and Barduhn, Allen J. "Recent Advances in Odor Control
By Air Washing," Annals of the New York Academy of Sciences, CXVI,
Art. 2 (July, 1964), 638-651.
The historical and practical basis of odor removal by scrubbing air
with liquids is considered. A merit factor system of evaluation is
advanced for determining total operating costs of this method.
Bretcher, Hans. "Elimination of Malodorous Emissions in the Chemical
Industry," Chemical Engineering, XLVI, No. 7 (September, 1968)
268-270.
Measures for control of malodorous emissions in the chemical industry
are discussed. In addition, areas such as meteorology, chromatography,
and process engineering methods are recommended for intensive future
work in relation to such emissions.
-238-
-------
Brewer, G. L. "Odor Control for Kettle Cooking, " Journal of the Air Pollution
Control Association, XIII (April, 1963), 167-169.
This paper discusses the odor problem of the paint and varnish industry,
and explains the catalytic system of odor control.
Brooman, David L., and Edgerly, Edward J. "Concentration and Recovery of
Atmospheric Odor Pollutants Using Activated Carbon, " Journal of the
Air Pollution Control Association, XVI (January, 1966), 25-38.
Results of a series of removal studies using activated carbon filters are
presented. These studies indicated that the filter capacity is dependent
on the type of carbon employed and the particular odor compound absorbed.
Byrd, J. F., Mills, H. A., Schellhose, C. H., and Stokes, H. E. "Solving a
Major Odor Problem in a Chemical Process, " Journal of the Air Pollution
Control Association, XIV (December, 1964), 509.
This paper describes the control of a highly objectionable odor from a
chemical raw material to be used in synthetic detergents. It also pro-
vides some guidelines for odor control in a complex process.
Chass, Robert L., Kanter, CarlV., and Elliott, JackH. "Contribution of Solvents
to Air Pollution and Methods for Controlling Their Emissions, " Journal of
the Air Pollution Control Association, XIII, No. 2 (February, 1963),
64-71.
A general discussion of solvent injection into the air over Los Angeles
County and procedures for its control are presented. Control by ad-
sorption was selected for the experimental work.
Douglass, Irwin B. "Some Chemical Aspects of Kraft Odor Control, " Journal of
the Air Pollution Control Association, XVIII (August, 1968), 541-544.
The principal sources of odor in the kraft pulping process, methods of
control, and the chemistry of various control measures are discussed.
-239-
-------
Fink, R. H. "A Management Approach to Odor Control, " Journal of the Air
Pollution Control Association, LVIII (May, 1957), 57-58.
Fundamental key steps for management's approach to odor control,
regardless of the industry or company, are provided.
Fohl, Timothy. "Optimization of Flow for Forcing Stack Wastes to High Alti-
tudes, ''Jo^iniaJU>fJfte_Air_^ XVII, No. 11
(November, 1967), 730-732.
This article discusses a procedure of waste gas projection to relatively
high altitudes. By this method, pollution products are then carried to
regions of low population density.
Gommi, J. V. "Design and Operation of Ace System for Kraft Odor Reduction, "
Paper Trade Journal. CLIV (July 6, 1970), 44-46.
Pilot development and the first commercially installed air contact evapo-
rator system which eliminates the kraft recovery furnace contribution
to total reduced sulfur emission are discussed.
Hanna, G. F., Kuehner, R. L., Karne, J. D,, and Garbowicz, R. "A Chemical
Method for Odor Control, " Annals of the New York Academy of Sciences,
CXVI, Art. 2 (July, 1964), 663-675.
An air deodorant substance using potassium permanganate in combination
with activated substances is described.
Hansen, G. A. "Odor and Fallout Control in a Kraft Pulp Mill, " Journal of the
Air Pollution Control Association, XII (September, 1962), 409-413.
The scope of the kraft mill pollution problem is discussed and the odor
control methods used by one company are presented.
Hein, Glen M. "Odor Control by Catalytic and High-Temperature Oxidation, "
Annals of the New York Academy of Sciences, CXVI, Art. 2 (July, 1964)
. 656-662.
This paper describes direct-flame and catalytic combustion as applied
to odor control, and discusses operating costs and new developments .
-240-
-------
Henshow, T. B. "Odor Control at a 2, 4-D Production Plant, " Journal of the Air
Pollution Control Association, XV, No. 11 (November, 1965), 516-518.
This article presents a discussion of the Chipman Chemical Company's
system of fume collection and caustic soda solution scrubbing for the
removal of phenolic compounds in plant exhaust air. The development,
present status, and projected future improvements of this system of
odor control are discussed.
Herrick, Robert A. "A Baghouse Test Program for Oxygen Lanced Open Hearth
Fume Control, " Journal of the Air Pollution Control Association, XIII,
No. 1 (January, 1963), 28-32.
A discussion of a pilot study of a baghouse type dust collector to deter-
mine its possible merit in the control of pollution from an oxygen lanced
open hearth furnace is presented. The results indicated a positive
collection of iron oxide fumes.
Ito, Shozo. "No Pollution or Catalyst In Continuous Oil Gasification, " Chemical
Engineering, LXXVIII (December, 1970), 113-115.
A Japanese steam reforming technique which does not require a catalyst
or carbon monoxide converter is described. A typical analysis of pro-
duct gas is given.
Japikse, B. "Threshold Approach to Estimating Adsorber Performance, " Annals
of the New York Academy of Sciences, CXVI, Art. 2 (July, 1964), 652-
655.
An approach to estimating performance of activated charcoal adsorber
cells used in air purification based on the concept of odor thresholds is
advanced.
Jones, K. H., Thomas, J. F., and Brink, D. L. "Control of Malodors From
Kraft Recovery Operations by Pyrolysis," Journal of the Air Pollution
Control Association, XIX (July, 1969), 501-504.
A simplified odor model is presented, which establishes the constraints
which must be placed on the combustion phase of the recovery operation
of kraft mills in order that emission concentrations are kept below their
threshold odor levels.
-241-
-------
Kellum, James. "Sewer Odor Control, " Water and Waste Engineering, VII,
No. 7 (July, 1970), 43.
Advantages and equipment for a newly developed process for the control
of odors in sewers, lift stations, and sewage treatment plants is dis-
cussed.
Kulka, Kurt. "Odor Control By Modification, " Annals of the New York Academy
of Sciences, CXVI, Art. 2 (July, 1964), 676-681.
This paper is concerned with air freshening in confined spaces.
Larsen, Ralph I. "Determining Reduced Emissions Goals Needed to Achieve
Air Quality Goals: A Hypothetical Case, " Journal of the Air Pollution
Control Association, XVII, No. 12 (December, 1967), 823-828.
Discussion and explanation of a hypothetical case for the determination
of a set of reduced emission goals for stationary and mobile combustion
sources is provided.
Moorman, Robert, Jr. "Controlling Odors From Cattle Feed Lots and Manure
Dehydration Operations, " Journal of the Air Pollution Control Association,
XV (January, 1965), 34-35~
A study of various types of odor control measures observed from inspec-
tions of cattle feeding lots across the country is discussed. The dehydra-
tion method for storing manure prior to its dehydration is described.
Nelson, F., and Shenfeld, L. "Economics, Engineering and Air Pollution in the
Design of Large Chimneys, " Journal of the Air Pollution Control Associa-
tion, XV, No. 11 (November, 1965), 536-538.
A description of the method for chimney selection and design is provided.
A reduction in SO2 was observed due to large mass flow from a single
source and buoyancy effect.
-242-
-------
Ottoson, David G. R. "Maintenance of Odor-Control Systems, " Environmental
Control Management (1970), 58-59.
To prevent odor during fermentation from being vented into the atmos -
phere, engineers at a pharmaceutical company producing antibiotics
devised a system for conducting the fumes of the fermentation facilities
to the power plant for incineration. This system is discussed.
Persson, Goran A. "Odour and Air Quality Management. Paper presented at
the Conference on Methods for Measuring and Evaluating Odorous Air
Pollutants at the Source and in the Ambient Air, Stockholm, June 1-5,
1970.
This paper reviews some of the air pollution problems caused by odors;
hygienic, sanitary, and technological standards for air quality; practical
use of odor measurements; and some problems and questions for further
study.
'Pollution Control Progress," Journal of the Air Pollution Control Association,
XVII, No. 3 (March, 1967), 172-174.
A discussion of the active steps being taken by individual corporations
and individual plants in the battle against air pollution is presented.
Roberson, James E. "The Effect of Odor Control On a Kraft Mill Energy
Balance," Journal of the Air Pollution Control Association, XX (June,
1970), 373-376.
Thirteen different recovery unit odor control cases are analyzed to
determine the variation in energy costs. Control methods such as
oxidation and new recovery systems are compared for a new 600 ton/
day kraft pulp mill.
Rowe, Norman R. "Odor Control With Activated Charcoal, " Journal of the Air
Pollution Control Association, XIII (April, 1963), 150-153.
A discussion of effective, activated charcoal, odor-control systems is
presented.
-243-
-------
Ryason, P. R., and Harkins, J. "Studies On a New Method of Simultaneously
Removing Sulfur Dioxide and Oxides of Nitrogen From Combustion Gases, "
Journal of the Air Pollution Control Association, XVII, No. 12 (December,
1967), 796-799.
A substantial reduction of sulfur compounds and oxides of nitrogen in
flue gases is possible by catalyzed reaction with carbon monoxide.
The economics of this process are estimated to be similar to those
of the catalytic oxidation process.
Stern, Arthur C. "Summary of the Conference on Odor Control, " Journal of the
Air Pollution Control Association, LVIII (May, 1957), 53-84.
This paper summarizes the information presented at the Symposium on
Odor Evaluation and Odor Control, Cincinnati, Ohio, October 25, 1956.
Strauss, W. "The Development of a Condenser for Odor Control From Dry Ren-
dering Plants," Journal of the Air Pollution Control Association, XIV
(October, 1964), 424-426.
This paper discusses a surface condenser and carbon absorber which
provides an economic solution to the problem of odors from dry ren-
dering plants.
Sullivan, J. L., Kafka, F. L., and Ferrari, L. M. "An Evaluation of Catalytic
and Direct Fired Afterburners for Coffee and Chicory Roasting Odors, "
Journal of the Air Pollution Control Association, XV, No. 12 (December,
1965), 583-586.
This paper discusses a method of using thermal afterburners in the
reduction of odor from large scale plants. Where high cost of fuel
prohibited the use of thermal afterburners, catalytic combustion was
used as an alternative.
Sweeney, Patrick M. "Exhaust Control Devices, " Journal of the Air Pollution
Control Association, XV, No. 1 (January, 1965), 13-14.
This article summarizes the developments in the area of exhaust con-
trol up to 1965.
-244-
-------
Teller, A, J. "Odor Abatement in the Rendering and Allied Industries, " Journal
of the Air Pollution Control Association, XIII (April, 1963), 148-149.
A review of operational odor abatement systems designed for the three
different types of rendering operations is presented.
Tucker, W. Gene. "Odor Control Technology in the Wood Pulping Industry."
Paper presented at the Conference on Methods for Measuring and Evalu-
ating Odorous Air Pollutants at the Source and in the Ambient Air,
Stockholm, June 1-5, 1970.
The approximate limiting capability of present odor control technology
for kraft pulp mills is discussed in terms of minimum overall emissions
of total reduced sulfur compounds. Costs of such control are estimated,
maximum ambient ground level concentrations are calculated, and factors
in determining the types of odor control research and development that
should be undertaken in the future are discussed.
Turk, Amos. "Odor Control," Encyclopedia of Chemical Technology, XIV (New
York: John Wiley & Sons, Inc., 1967), 170-178.
This article reviews the various methods of odor control and gives
some consideration to the relative effectiveness and cost of these
methods.
Vedernickov, V. G., and Maksimov, V. F. "Some Problems in the Deodoriza-
tion of Gas Discharges in the Sulf ate-Cellulose Industry." (Franklin
Institute Research Laboratories, Philadelphia, Pennsylvania, Science
Information Services, November 11, 1969).
This article discusses the use of uncondensed exhaust gases and deodor-
izing agents for the sulf ate-cellulose industry.
Viessman, Warren. "Ventilation Control of Odor, " Annals of the New York
Academy of Sciences, CXVI, Art. 2 (July, 1964), 630-637.
This article includes a discussion of odors, odor sources, and ventila-
tion as a means of odor control.
-245-
-------
Walker, A. B. "Operating Principles of Air Pollution Control Equipment. Guide-
lines for Their Application, " Proceedings of the MeCAR Symposium,
Design and Operation for Air Pollution Control (American Institute of
Chemical Engineers, New York, New York, 1968), 49-74.
The basic operating principles of equipment for emission control are
examined and the critical factors in their application are identified.
-246-
-------
E. ODOR CONTROL ORDINANCES
Earth, D. S. "Administrative Mechanisms Available for Control of Odorous
Compounds Under the U.S. Clean Air Act, As Amended." Paper pre-
sented at the working conference on The. Dose-Response Relationships
Affecting Human Reactions to Odorous Compounds, a symposium spon-
sored by the Environmental Protection Agency, Air Pollution Control
Office, Cambridge, Massachusetts, April 26-27, 1971.
Some of the authorities granted to the Administrator, Environmental
Protection Agency, by the amended Clean Air Act are discussed in
terms of their possible applicability to the control of odorous compounds
Barynin, Jan. "Measuring Odour Pollution, " New Scientist, XLVIII, No. 723
(October 15, 1970), 116-119.
The prospects of introducing scientific standards for assessing odor
pollution as an aid towards legislation on odor abatement are discussed.
Comparison of the photomultiplier tube and observer responses are also
considered.
Bistowish, Joseph M. "Nashville's Air Pollution Problem: Past, Present, and
Future." Vanderbilt University, Nashville, Tennessee, School of
Engineering, Tennessee Stream Pollution Control Board, Tennessee
Department of Public Health, Proc. Conf. Environ., Water Resources
Engineering, 85th Anniversary, Nashville, Tennessee, 1969, 171-177.
The. provisions and implementation of an air pollution control ordinance
prepared by the Nashville Department of Health, including provisions
for odor pollution control, are discussed.
"Control of Air Pollution, " International Digest of Health Legislation, XX, No. 3
(1969), 418-419.
The principal provisions of a regulation under an air pollution control
act for Ontario are presented. The regulation states that equipment
may not be operated which does not comply with minimum air quality
standards or emits disagreeable odors. Maximum allowable concen-
trations for use in assessing air quality are also prescribed.
-247-
-------
Cross, Frank L., Jr., and Ross, Roger W. "Current Status of Air Pollution
Control Regulations for Asphalt Plants and Similar Drying Operations."
Preprint, Public Health Service, Durham, North Carolina, National Air
Pollution Control Administration, September, 1969.
Current regulations for controlling emissions from asphalt plants are
reported for 38 states and 6 local agencies. Nine states employ a stan-
dard for odor control. Most of the odor regulations are general in
nature, although one state stipulates the use of a scentometer for odor
measurement.
First, M. W. "A Model Odor Control Ordinance." Paper presented at the
Second International Clean Air Congress of the International Union of
Air Pollution Prevention Associations, Washington, D. C,,
December 6-11, 1970.
A discussion of the current pollution regulations concerned with odor,
odor perception and evaluation, and odor nuisance evaluation is pro-
vided along with a model odor control ordinance that provides a numer-
ical method for arriving at an objective judgment of the degree of
nuisance associated with a particular set of conditions.
Gruber, Charles W. "Odor Pollution Problems From the Control Officials'
Viewpoint," Journal of the Air Pollution Control Association, LVIII
(May, 1957), 56-57.
A discussion of odor pollution from the control officials' point of view is
provided. Topics included are: Finding the Odor Source, Evaluating the
Odor Situation, What Can Be Done About an Odor Situation, and Odor
Monitoring Runs.
Jonsson, Erland. "Nuisance From External Environmental Factors and Norms
for Their Evaluation," Nordisk Hygienisk Tidskrift, XLIV (1963), 69-84,
A summary of the background and results of four studies on problems
connected with the decisions about the presence of "sanitary nuisance"
is given. Suggestions are made for measures which would increase the
likelihood of correct and well-founded decisions being made as to the
presence of a "sanitary nuisance."
-248-
-------
Larsen, Ralph I. "Relating Air Pollutant Effects to Concentration and Control."
National Air Pollution Control Administration, Durham, North Carolina,
Office of Criteria and Standards, 1969.
Examples are cited relating air pollutant effects to concentration and
control. The examples are organized into a dozen possible steps that
could be used to proceed from air quality criteria to clean air.
Lindemann, E. H. "Odor Control — 1953," Annals of the New York Academy of
Sciences, CXVI, Art. 2 (July, 1964), 623-629.
The state of the art of odor control in 1953 is surveyed. Methods of
eliminating odors by mechanical or physical means, chemical means,
and by masking or modification are presented. An extensive biblio-
graphy of the literature on the subject in the late 1940's and early
1950's is also presented.
Stumph, Terry L., and Duprey, Robert L. "Trends in Air Pollution Control
Regulations," Environmental Control Management, CXXXIX, No. 2
(February, 1960), 45-46.
Some existing odor control regulations and their partially successful
measures are discussed. Possible developments in control regulations
are also presented.
"Summary of Air Pollution Actions in 1966 State Legislatures, " Journal of the
Air Pollution Control Association, XVII, No. 2 (February, 1967), 115-
118.
A brief analysis of the legislation concerning air pollution in 1966 is
presented.
Turk, Amos. "A Basis for a Model Ordinance to Control Effectively Commu-
nity Odor Problems." Report submitted to the U. S. Public Health
Service, Contract No. PH 27-65-104, September 29, 1965.
This report provides a basis for a model ordinance and covers such
topics as points of measurements, methods of measurements, allow -
able concentration limits, and experimental test programs necessary
for the implementation of the methods suggested in the report.
-249-
-------
Walsh, Robert T. "The Bishop Case — An Interstate Odor Problem." Preprint,
National Air Pollution Control Administration, Durham, North Carolina,
Division of Abatement, 1970.
A discussion of the technical evaluation, the abatement conference,
recommended remedial measures, and ensuing public hearing against
one company is presented. The company's failure to carry out the
remedial measures led to a suit in the U.S. District Court in Maryland
and the eventual order to shut the plant down.
-250-
-------
F. GENERAL INFORMATION ON ODOR POLLUTION
Gonzalez, D. J. "Odor Problems: A Review," Annals of the New York Academy
of Sciences, CXVI, Art. 2 (July, 1964), 621-622.
A brief review of some odor problems in the early 1950's is presented.
Research Triangle Institute, Research Triangle Park, North Carolina, Engineer-
ing and Environmental Sciences Division. Applications of Aerospace
Technology in Air Pollution Control. Final Report. NASA Contract
NSR-34-004-056, RTIUE-411-3, June, 1969.
This paper is a report of the work of the Technology Application Team
of Research Triangle Institute which identifies very specific technology-
related problems and prepares air and odor pollution problem abstracts
which describe such problems in a concise manner using functional,
nondisciplinary terminology.
Rose, A. H., Smith, R., McMichael, W. F., and Kruse, R. E. "Comparisons
of Auto Exhaust Emissions in Two Major Cities, " Journal of the Air
Pollution Control Association, XV, No. 8 (August, 1965), 362-366.
A detailed discussion of the comparison of auto exhaust in a low-altitude
city to auto exhaust in a high-altitude city is presented.
Sullivan, Ralph J. "Preliminary Air Pollution Survey of Odorous Compounds."
A Literature Review. Litton Systems, Inc., Silver Springs, Maryland,
Environmental Systems Division, Contract PH 22-68-25, NAPCA Publ.
APTP 66-42, October, 1969.
A review of the literature concerning the effects, sources, abatement
methods, and detectors of odorous pollutants is presented.
Suzuki, Takeo. "Summary of the Problems in Odorous Air Pollution in Japan."
Paper presented at the working conference on The Dose-Response
Relationships Affecting Human Reactions to Odorous Compounds, a sym-
posium sponsored by the Environmental Protection Agency, Air Pollution
Control Office, Cambridge, Massachusetts, April 26-27, 1971.
This paper reviews the odor problem in Japan and covers such topics as
features of the odor problem, major malodorous compounds and their
sources, reactions of the people against odor, and odor control.
-251-
-------
PROCEDURE MANUAL
-253-
-------
PROCEDURES FOR THE
IDENTIFICATION AND ASSESSMENT
COMMUNITY ODOR PROBLEMS
Prepared for the
ENVIRONMENTAL PROTECTION AGENCY
Under Contract No. CPA 70-116
By
COPLEY INTERNATIONAL CORPORATION
7817 Herschel Avenue
Lajolla, California 92037
Contributions were made to this manual by
POPE, EVANS AND ROBBINS
564 Market Street
San Francisco, California 94104
November 1971
-------
ACKNOWLEDGMENTS
Copley International Corporation gratefully acknowledges the cooperation
of many private citizens who gave generously of their time by participating in
personal interviews. Without their wholehearted cooperation, this manual would
not have been possible. Copley International Corporation is especially grateful
to Mr. Richard C. Dickerson of the Environmental Protection Agency for his
counsel and assistance in the development of this document.
CONTRIBUTIONS
The overall responsibility of this study was undertaken by Mr. R . David
Flesh, Director, Environmental Economics, Copley International Corporation.
Others who contributed to this manual included:
Dr. Amos Turk, Project Consultant and Professor, City College of the
City University of New York.
Mr. James C. Burns, Senior Staff Chemist; and Mr. P.M. Conn, Vice
President; of Pope, Evans and Robbins .
Mrs. Marian O. Doscher, Senior Industrial Economist; and Mr.
Thomas H. Copeland, Director, Marketing and Behavioral Research; of Copley
International Corporation.
-i-
-------
TABLE OF CONTENTS
ACKNOWLEDGMENTS AND CONTRIBUTIONS . i
LIST OF TABLES v
LIST OF FIGURES v
FORMS vi
IDENTIFICATION AND ASSESSMENT OF COMMUNITY ODOR
PROBLEMS 1
INTRODUCTION 1
BACKGROUND INFORMATION 1
CONDITIONS WARRANTING USE OF THE PROCEDURES .... 3
PROCEDURE FOR PROBLEM IDENTIFICATION BASED ON BRIEF
PUBLIC ATTITUDE SURVEYS . . , . 5
INTRODUCTION 5
WHEN TO USE PUBLIC ATTITUDE SURVEYS 6
SETTING UP TEST AND CONTROL AREAS . 6
OBTAINING A SAMPLE ......... 7
CONDUCTING THE SURVEY 8
PROBLEM IDENTIFICATION 15
OTHER USES OF SURVEY RESULTS 16
STATISTICAL REFERENCE SECTION 26
PROCEDURE FOR SOURCE VERIFICATION BASED ON SENSORY
TECHNIQUES 31
INTRODUCTION 31
-111-
-------
TABLE OF CONTENTS (Cont'd)
WHEN TO USE THE SCENTOMETER 31
WHEN TO USE AN ODOR JUDGMENT PANEL 32
THE SCENTOMETER 32
THE ODOR JUDGMENT PANEL 35
SOURCE VERIFICATION 41
PROCEDURE FOR PROBLEM ASSESSMENT 47
INTRODUCTION 47
WHEN TO ATTEMPT PROBLEM ASSESSMENT 47
APPENDIX 49
ESTIMATES OF COSTS INCURRED FROM USING
PROCEDURES 49
SUGGESTED REFERENCES 54
-iv-
-------
LIST OF TABLES
Page
Sampling interval 8
Concentration of reference standard in odorless solvent
used to simulate four strengths of the odor expected
to be encountered in the test area 38
Recommended work schedule 40
LIST OF FIGURES
Figure Page
1 Barnebey-Cheney Model 1-3 Scentometer 33
2 Example of isopleths representing mean odor intensities
recorded by odor judgment panelists stationed in a
test area at the points indicated 43
3 Example of isopleths representing percentages of time
odor was detected by odor judgment panelists sta-
tioned in a test area at the points indicated 43
-v-
-------
FORMS
PUBLIC ATTITUDE SURVEY OF COMMUNITY ODOR PROBLEMS
PROBLEM IDENTIFICATION QUESTIONNAIRE 17
PUBLIC ATTITUDE SURVEY — TABULATION FORM 21
ODOR PROBLEM IDENTIFICATION 24
DETERMINATION OF THE ODOR PROBLEM INDEX 25
SCENTOMETER LOG SHEET 44
ODOR INTENSITY RATING SHEET 45
»
ODOR INTENSITY RATING SUMMARY SHEET 46
-vu-
-------
IDENTIFICATION AND ASSESSMENT OF
COMMUNITY ODOR PROBLEMS
INTRODUCTION
The procedures described in this manual are designed to assist govern-
mental agencies in dealings with community odor problems caused, particularly,
by industrial operations. The procedures are addressed to "local" agencies
since the investigation of community odor problems usually occurs within the
public health departments or air quality offices of local government. They are,
however, equally applicable to the needs of state agencies operating at the local
level.
BACKGROUND INFORMATION
The contents of this manual were developed primarily from the results of
two studies — the national survey of the odor problem, conducted in 1969, and a
study of the social and economic impact of odors, conducted in 1970 and 1971.
Both were conducted by Copley International Corporation for the Environmental
Protection Agency. Secondarily, the contents evolved from the findings of other
research done in the United States and Sweden, and from the opinions of scien-
tists, jurists, and many local agency officials. The procedures represent a logi-
cal sequence of investigation from problem identification and source verification
to a promise of uncomplicated methods of social and economic assessment.
The difficulty with the investigation of community odor problems lies in
the lack of objective evidence that such problems exist. Analytical equipment
is available to measure low concentrations of a few odorous substances, but
even in situations involving such substances, there is no way to translate known
concentrations into odor intensities. Equally distressing, there is no way to
translate known odor intensities into an index of problem perception. Thus, for
the present at least, reliance must be placed on personal evaluation by the normal
or ordinary person in the community.
Traditional methods for solving community odor problems include the
presentation of subjective evidence to the courts for consideration under public
nuisance law, which, for remedy, requires substantial and unreasonable inter-
ference to the normal person in the community, The most widely used form of
such evidence has been complaints initiated by residents. On the surface, it
would seem that such evidence must define the existence of a problem. Yet,
under a literal interpretation of public nuisance law, it may not. It is seldom
determined from complaints whether the complainants, individually or as a
group, actually represent the ordinary person or whether the ordinary person
has suffered interference.
-1-
-------
The procedures described in this manual are offered for use in connection
with public nuisance law in lieu of counting complaints or attempting to extract
meaningful information from complainants' remarks. Within these procedures,
complaints are relegated to a proper function of alerting local authorities to the
locations of possible odor problems.
Problem Identification Based on Public Attitude Surveys
The most direct means of identifying community odor problems is an
attitude survey. This stems from the personal nature of odor evaluation and
the concept that attitudes reflect personal evaluations. Unlike odor complaints,
an attitude survey of a randomly selected sample of residents would satisfy pub-
lic nuisance law. The random sample would represent the normal person. Atti-
tudes expressing annoyance (typical in odor cases) would represent interference.
To determine whether interference is substantial and unreasonable creates
a complexity. Equity is involved. It is unfair to find substantial and unreasonable
interference on the basis of an arbitrary percentage of residents who express
annoyance. Recent studies have found that, even in virtually odor free areas,
up to 50 percent of the residents may express annoyance to odors. This is due
to a variety of factors including, for example, residents with unusual sensitivity
to background odors, feelings against nearby industry, and neurosis.
The procedure for problem identification provides an equitable basis for
showing the extent of interference suffered. It is assumed that the percentage of
residents who would express annoyance to odors when odors are not present is
approximately equal in all communities of similar socioeconomic characteristics
in a particular metropolitan area. (Once established for communities of similar
socioeconomic characteristics, the percentages must not be applied to commu-
nities of other characteristics in the same metropolitan area or to communities
of similar characteristics in other metropolitan areas. This assumption is
supported by the results of the study of the social and economic impact of odors.)
It is also assumed that residents of all communities have equal right to odor free
air. By comparing the results of identical attitude surveys conducted in a sus-
pected odor problem community and a similar, but odor free community, it can
be statistically determined if a greater percentage of residents in the former
area express annoyance. If so, an odor problem is identified and an equitable
basis for showing substantial and unreasonable interference is provided.
Source Verification Based on Sensory Techniques
Whenever an odor problem has been identified, the source of the odors
must be found if the problem is to be remedied. When odors are experienced
in a community frequently or over long duration, it is common for the residents
to know or at least suspect the source. Still, the source must be verified by
local authorities.
-2-
-------
Source verification is a straightforward endeavor when the odor is easily
recognized and only one source capable of causing such an odor is located near
the problem community. In such cases, a simple investigation is adequate. How-
ever, when the odor is not familiar or occurs downwind of more than one possible
source, the specific offender may be difficult to pinpoint. To deal efficiently with
such cases, the use of sensory techniques must be considered. The procedure
for source verification describes two of the most successful techniques — one
featuring immediate availability, the other, depth of evaluation.
The use of sensory techniques in problem identification may be possible
in the future. This alternative to the use of public attitude surveys may save time
(including the time of residents who would otherwise be interviewed) and expense
without much loss of confidence in the results. What is needed to examine this
possibility is concurrent use of the sensory techniques and public attitude surveys
recommended by this manual on numerous occasions over time.
Based on the outcome of field tests performed by its developers, the scent-
ometer is recommended for indicating the locations of possible odor problems
during routine surveillance. A condition for its use in this capacity is given in a
following paragraph.
Problem Assessment
By problem assessment is meant a determination of economic and social
impact of odors on a community. Such information would support, but not replace
the need for problem identification in odor cases. Tentative methods of problem
assessment have been developed, the most comprehensive of which are documented
in the final report of the social and economic impact study. However, because
they are not yet refined to the level of problem identification and source verifi-
cation and because they require highly specialized social science skills not
possessed by most local agencies, they are not included in this manual.
CONDITIONS WARRANTING USE OF THE PROCEDURES
Two types of odor situations may exist in a community. One may lead to
an odor problem; the other does not. They are referred to as acute episodes and
chronic situations and are distinguished under public nuisance law, which is not
concerned with occasional causation of odors. For purposes of the following
definitions, causation of odors is meant to infer continued emission of odorous
substances into the atmosphere, not merely a result of impaired atmospheric
dilution capacity.
-------
Acute Episodes
An acute episode is defined as a causation of odors beyond the property
limits of a source, occurring no more than once and lasting no more than one
day in any three month period. An acute episode involving the emission of mea-
surable concentrations of odorous substances into the atmosphere should be dealt
with as a possible infraction of the air quality standards governing such substances.
An acute episode should not be considered as a possible odor problem regardless of
the characteristics of the odors caused.
Chronic Situations
A chronic situation is defined as a causation of odors beyond the property
limits of a source occurring more than once or lasting more than one day in any
three month period. A chronic situation involving the emission of measurable
concentrations of odorous substances should be dealt with as a possible infraction
of the air quality standards governing such substances. A chronic situation should
be considered as a possible odor problem under any of the following conditions:
(1) If one or more odor complaints are initiated by residents
of a community and verified by local authorities after the
first day of the first occurrence in any three month period.
(2) If twenty or more odor complaints are initiated by resi-
dents of a community, but not verified by local authorities
after the first day of the first occurrence in any three
month period.
(3) During routine surveillance by local authorities when trav-
eling through a community, if odors are detected on more
than one day in any three month period. Such odors must
be of sufficient intensity to be detected using a Barnebey-
Cheney Model 1-3 Scentometer set at 7 dilutions to threshold.
The existence of a chronic situation and any of these conditions should be neces-
sary and sufficient for the employment of the following procedures.
-4-
-------
PROCEDURE FOR PROBLEM IDENTIFICATION
BASED ON BRIEF PUBLIC ATTITUDE SURVEYS
INTRODUCTION
The following procedure is designed to allow local agencies to conduct,
with their own personnel, public attitude surveys to identify community odor
problems . If sufficient funds are available to the local agency, a professional
interviewing or survey firm could be engaged to perform many of the tasks de-
scribed in this procedure. However, the additional expense is not necessary if
the procedure outlined is followed with care by local agency personnel. It is
important to note that failure to adhere to the procedure could lead to survey
results that are misleading or meaningless .
Before describing the detailed steps to be undertaken in conducting a
public attitude survey, it may be helpful to summarize the major tasks to be
completed. These tasks are described in general terms below.
This survey is designed to compare the attitudes of people residing in a
community thought to be an odor problem area with attitudes of similar people
residing in an odor free area. Attitudes of both groups are determined by con-
ducting interviews by telephone with residents of both areas. (Tasks followed
by an asterisk (*) could be done by an interviewing firm.)
(1) The first task is to define the possible odor problem area.
(2) Next, a matching odor free area is located.
(3) Utilizing a street address (reverse order) telephone direc-
tory, a list of telephone numbers in each area is made, and
a sample of these telephone numbers is selected at random.*
(4) Utilizing the questionnaire provided in this procedure, tele-
phone interviews are conducted with the man or lady of the
house for each telephone number included in the sample. *
(5) The total number of responses to key questions asked in both
areas is then tabulated and compared for problem identifica-
tion.
(6) Finally, if an odor problem is found, an odor problem index
number is calculated.
-5-
-------
WHEN TO USE PUBLIC ATTITUDE SURVEYS
A public attitude survey should be conducted only after any of the three
conditions stated in the previous section of this procedure indicate the location
of a possible odor problem.
SETTING UP TEST AND CONTROL AREAS
The community in which an odor problem is suspected is called the "test
area." The first task to be undertaken is to define the preliminary geographic
limits of the test area. This can be done by driving an automobile in a grid
pattern throughout the general area, while noting the geographical boundaries
within which the odor is perceived. Meteorological data indicating wind patterns
can also be helpful in setting these boundaries, as can complaint patterns. It is
necessary to establish a test area that has boundaries coincident with those of
census tracts. If only a portion of a census tract is included in the preliminary
boundaries of the test area, that census tract should not be used unless a large
proportion of the entire census tract is probably affected by odors . It is impor-
tant to include only those tracts where an odor problem is likely; otherwise, the
results of the survey may not be conclusive.
Next, the socioeconomic characteristics of the test area must be deter-
mined. Using the latest U.S. Censuses of Population and Housing Census Tract
Reports (PHC), list the following characteristics of the test area: median income
for families, median value of owner occupied housing units, median gross rent
of renter occupied units, median number of rooms, and year structure was built.
An odor free community must now be located that matches the test area
as closely as possible. This community is called the "control area." The
following criteria should be used in choosing the control area.
(1) The control area should be as odor free as possible. To
ensure this, the local agency should not have received any
odor complaints from residents of the area during the past
year.
(2) The control area should be located within ten miles of the
test area..
(3) The control area should have the same type of access to
freeways or turnpikes as the test area.
(4) The control area should be located within approximately
the same distance from commercial or industrial estab-
lishments as the test area.
-6-
-------
(5) The median income, home value, gross rent, and age of
structure (the latter must be figured from year structure
was built) for the control area should not differ from that
of the test area by more than 20 percent.
(6) Median rooms (per house) should not differ by more than
10 percent for any one category. For example, suppose
the test area contains 646 housing units having five rooms
out of a total of 2,332 housing units; then, 27 .7% are in
the five room category. Suppose the control area contains
375 units having five rooms out of 1,981 units; then, 18 .9%
of the units are in the five room category. For this exam -
pie, the difference is 8.8% (27.7% - 18.9% = 8.8%), which
is acceptable.
OBTAINING A SAMPLE
First, list all the streets in the census tract(s) that have been selected for
the test area. Note the house number of the first and last house within the census
tract(s) on each street that runs beyond the test area. Using a street address
(reverse order) telephone directory, which can be obtained from the telephone
company, count the total number of telephones that are listed for streets in the
test area. In this count, do not include professional offices, commercial estab-
lishments, government offices, or industrial facilities . Next, using a table of
random numbers, select one number that falls between the number "1" and the
total number of telephones in the test area. Count from the first test area address
in the directory to this number. This is the starting point, i.e., the first tele-
phone number to be used for the sample. Refer to Table 1 to determine the num-
ber of telephone numbers to be passed over before a second telephone number is
chosen (the sampling interval). Write down this address and telephone number.
Continue in this manner until all telephone numbers in the test area have been
written down or passed over as part of the sampling interval. It should be noted
that if the starting point is toward the middle of the alphabetically listed streets
in the directory, when the last street in the test area is reached, the sampling
should be continued by returning to the first alphabetically listed street in the
test area.
When this procedure has been completed for the test area, the same thing
should be done for the control area.
-7-
-------
Table 1. Sampling interval.
Number of Telephones
in the Area
1 - 90
91 - 180
181 - 270
271 - 360
361 - 450
451 - 540
541 - 630
631 - 720
721 - 810
811 - 900
901 - 990
911 - 1080
1081 - 1170
1171 - 1260
1261 - 1350
1351 and over
Sampling Interval
include all telephones
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
CONDUCTING THE SURVEY
Each time interviews are conducted in the test area, they must also be
conducted in a matching control area. This is necessary to permit the statis-
tical comparisons required to verify the findings of the survey. Surveys should
be conducted concurrently in the test and control areas. The same control area
may be used for different test areas if the residents have the same socioeconomic
characteristics as those in a second test area.
Telephone calls to administer the questionnaire should not be made before
9:00 a.m. nor after 9:00 p.m. To reduce the elapsed time needed to complete
the telephone calls, the samples may be divided among two or more interviewers.
If only two interviewers are used, one may be assigned to call the control area
sample while the other calls the test area sample. To reduce the possibility of
interviewers biasing the results of the survey, it is desirable to switch the inter-
viewers about half way through the sample, i.e., after one-half of the telephone
calls in the test area have been made by one interviewer, this person should make
calls in the control area. The other interviewer will make the balance of the test
area calls. A blank copy of the questionnaire will be needed for each telephone
number to be called.
Pretesting the Survey
Because it is unlikely that local agency personnel are experienced in con-
ducting telephone surveys, it will be of considerable help to pretest the survey.
-8-
-------
However, the pretest results should not be combined with those of the actual
survey.
Arrange for the persons who will conduct the telephone interviews to
read the "Instructions for Interviewers" section of this procedure. When this
has been done, have each interviewer go through the questionnaire asking the
questions aloud of another agency employee. Note whether the interviewer is
following instructions in every detail, and correct any errors in interviewing
technique. Then, select at random five telephone numbers for each interviewer
who will participate in the survey. These telephone numbers should be differ-
ent from those chosen as part of either the test or control area samples. Have
each interviewer call five numbers and administer the questionnaire. Again,
note whether the interviewer is following instructions, and correct any errors.
When each interviewer is thoroughly familiar with the questionnaire and inter-
viewing technique, begin calling the telephone numbers in the test and control
area samples.
Instructions for Interviewers
Because people would usually like to tell you what they believe you want to
hear, it is important that the interviewer restrict what is said to the questions
or comments on the questionnaire. If the interviewer engages in pleasantries or
chats with the person answering the questions, the responses received may be
different from those that would otherwise be obtained. The interviewer should
be particularly careful not to indicate by either comment or tone of voice any
reaction to the answers received. It is also important for the interviewer not to
reveal the name of the agency conducting the survey nor the reason the survey
is being conducted. Some respondents may challenge the interviewer or question
the validity of the survey. Such situations must be handled with a great deal of
tact on the part of the interviewer. The respondent has the right to determine
whether the survey is legitimate, but if information about the agency or the sur-
vey is given before the questions are asked, the respondent's answers may not
be valid. Therefore, if possible, the interviewer should agree to tell the respon-
dent this information after the questionnaire is completed. If the respondent
refuses to continue the interview without this information, the interview should
be terminated.
The respondent may suspect that the call is being made in connection with
the sale of a product. The interviewer may assure the respondent that this is
not the case prior to asking the questions . It may also be necessary to tell the
respondent that the answers given will become a part of a statistical summary
and individual responses will not be revealed in any way. If required, this may
also be done prior to asking the questions .
Never suggest an answer or train of thought to the respondent. Do not
prompt in any way. The most important job of an interviewer is to encourage
-9-
-------
the respondent to express his or her attitudes freely and to be specific in con-
veying these attitudes. Close attention to the clarity and completeness of the
responses is important. Never talk up or down to a respondent. Do not com-
ment on the meaning of any question or indicate in any way what kind of answers
are expected.
If the respondent asks a question or fails to understand or respond to a
question, read the question over again more slowly, using a pleasant but neutral
tone of voice. On occasion, it may be necessary to repeat a question more than
once before obtaining a response. Do not allow any tone of impatience to creep
into your voice.
When asking the questions, stress only those words that are underlined.
Do not change the wording of any question. These questions have been carefully
constructed to obtain the information desired. Do not change the order of the
questions from that listed on the questionnaire.
Before making a telephone call, the interviewer should fill out the upper
portion of the questionnaire (a copy of the questionnaire is included at the end of
this procedure). From the sample listing, fill in the telephone number, street
address, and city to which the call will be made. The "File No." blank is pro-
vided for identification of the odor source under investigation. This blank should
not be filled in until the source has been verified. The "Survey No." blank is
provided to allow differentiation between test and control area responses, as well
as surveys conducted at a later time. For example, before the first survey of a
suspected odor problem, "Survey No. 1" could be assigned to the test area and
"Survey No. 2" could be assigned to the control area. If the same areas are the
subject of another survey at a later date, survey numbers 3 and 4 could be
assigned, respectively.
The "Respondent No." should be left blank until tabulation of the survey
results is begun. However, the date of the call and the time of day the call is
made should be noted on the questionnaire before the call is made.
Interviewer Instructions for Making Calls
Call the telephone number filled in on the questionnaire. If there is no
answer or the line is busy, go on to the next telephone number on the list.
If the call is answered, read the opening statement, saying your name
in the appropriate place. If the person who answers the telephone is obviously
a woman, use the feminine alternatives in the statement; if a man, the mascu-
line alternatives. Children, relatives, or friends living in or visiting the house-
hold should not be interviewed. In addition, no person under 18 years of age
should be interviewed. Make sure that the person with whom you are speaking
is the man or lady of the house. If he or she is not, try to talk with the man or
-10-
-------
lady of the house. If this cannot be done, terminate the call, check that the
interview was not completed, note the reason, and go on to the next telephone
number in the sample.
If the man or lady of the house refuses to be interviewed at all, note that
the interview was not completed and the reason. "Refused" is a sufficient expla-
nation .
The numbered blanks in the right-hand margin of the questionnaire are
to be used in tabulating the responses . Do not write in these spaces at this time.
Every question has a list of responses after it. Except for question 3, only one
answer should be checked. If the respondent gives more than one answer for a
question, ask which answer is the best answer to the question. When a question
is followed by three periods, such as, "How strong would you say these odors
smell? Would you say...?", read the answers provided in the order they are
listed, but do not read either the answer numbers or the response "Don't know."
When a question is not followed by three periods, such as, "When was
the last time you noticed odors in your neighborhood?", do not read the responses.
In this case, simply check the answer that corresponds to the respondent's reply.
When a question has a space provided for "Other answers," write down every-
thing the respondent says. If local agency secretaries are used as interviewers,
they should use only longhand to write down "Other answers."
Interview Instructions for Specific Questions
The following instructions should be adhered to in asking the specific
questions on the questionnaire.
Question 1. Do not read the responses. If the respondent answers "No, "
read the statement following the "No" response and terminate the interview.
Question 2. Do not read the responses . If the respondent answers "Yes, "
ask question 3. If the respondent answers "No" or "Don't know," ask question 4.
Question 3. This question should be asked only of those people who
answered "Yes" to question 2. Read the responses, except for "Don't know."
Check all answers that the respondent gives to this question.
Question 4. Do not read the responses. If the respondent answers "No,"
or "Don't know, " skip to question 12. If the respondent answers "Yes, " ask
question 5.
Question 5. Do not read the responses . Check the response that covers
a period most like that identified by the respondent, Repeat the question, if neces-
sary, to obtain a response, but do not read the possible responses.
-11-
-------
Question 6. Do not read the responses. Check the response that covers
a period most like that identified by the respondent. Repeat the question, if neces-
sary, to obtain a response, but do not read the possible responses.
Question 7. Read the responses rather slowly, pausing between each
possible response. Do not read "Don't know." Repeat the question and responses,
if necessary, to obtain a reply.
Question JL Do not read the responses. Check the response that covers
a period most like that identified by the respondent. Repeat the question, if neces-
sary, to obtain a response, but do not read the possible responses .
Question 9. Do not read the responses . If the respondent answers "No"
or "Don't know," skip to question 11. If the respondent answers "Yes," ask
question 10.
Question 10. Read the responses rather slowly, pausing between each
possible response. Do not read "Don't know." Repeat the question and responses,
if necessary, to obtain a reply.
Question 11. Do not read the responses. Clarify the response, if neces-
sary, by asking the respondent to explain more fully what is meant by the answer
given. For example, if the respondent says "Some factory, " clarification might
elicit the answer, "An oil refinery near my neighborhood." Write down other
source(s) identified by the respondent.
Next, read the statement preceding question 12 regarding the confidential
nature of the information being gathered.
Question 12. Do not read the responses. Clarify the responses, if neces-
sary, as explained under question 11. Write down the other company mentioned
by the respondent if the answer is not the source(s) under investigation.
Question 13. Do not read the responses. Clarify the response, if neces-
sary, as explained under question 11. Write down the other companies or employ-
ers mentioned if the answer does not include the source(s) under investigation.
Interviewer Instructions for Completing Interviews
Read the statement at the end of the questionnaire and terminate the inter-
view. Write your name in the space provided, and check the space for a com-
pleted interview on the first page of the questionnaire. Go on to the next telephone
number.
Interviewing should be continued until all of the telephone numbers in the
sample have been dialed once. Review the completed questionnaires, and remove
-12-
-------
those where the answer to either question 12 or question 13 indicates a member
of the household is employed by the odor source under investigation. This is rec-
ommended because of potential bias the respondent may have either in favor of or
against the employer. If the source has not been verified by an agency inspector,
this step should be ignored.
Next, count the number of completed questionnaires remaining for the
test area. If there are at least 30, the interviewing portion of the survey is
complete. If not, continue to interview, utilizing those telephone numbers where
no-answers or busy signals were received, until at least 30 questionnaires are
completed or until all no-answers or busy signals in the sample have been called
a second time. Do the same thing for the control area.
Tabulation
If the local agency has access to electronic data processing equipment,
and the number of responses in either the test or control area exceeds 90, con-
sult with those in charge of such equipment to set up procedures for tabulating
the survey. If such equipment is not available or the number of responses in
each area is small, hand tabulating is quite satisfactory.
Regardless of the tabulating procedure utilized, the completed question-
naires must be edited and coded. Editing should be done by persons other than
those who conducted the interviews. The purpose of this task is to insure that
all questions have been asked and answers recorded. Inconsistencies also can
be identified and corrected.
On each questionnaire there should be a response to all appropriate^ques-
tions . For example, if question 2 concerning people in the neighborhood com -
plaining about pollution has been answered with a "Yes," then question 3 dealing
with the kind of pollution people complain about should have also been answered,
even if the answer is "Don't know."
The following comments regarding each question on the questionnaire
may be used as a guide for editing. If editing uncovers errors or omissions,
it may be necessary to call the respondent again to clarify an answer which has
been incorrectly recorded or to eliminate an inconsistency.
Question 1. If the response is "No, " all other questions on the ques-
tionnaire should be unanswered.
Question 2. If the response is "Yes," there should be an answer to
question 3. If the response is "No, " there should be no response to question 3.
Question 3. There should be no response to this question if question 2
has a "No" reply.
-13-
-------
Question 4. If the response is "No" or "Don't know, " questions 5, 6, 7,
8, 9, 10, and 11 should not contain responses. If the answer to question 4 is
"Yes," questions 5, 6, 7, 8, 9, and 11 must have been answered, also, even if
the answer is "Don't know."
Question 5. There should be a reply to this question only if question 4
was answered "Yes ." If so, there must be a reply to this question.
Question 6. Same as question 5.
Question 7 . Same as question 5.
Question 8. Same as question 5.
Question 9. Same as question 5.
Question 10. If the response to question 9 was "No" or "Don't know,"
there should be no response to this question. If the response to both question 4
and question 9 was "Yes," this question should have been answered, also .
Question 11. Same as question 5.
Question 12. There should be a response to this question on all question-
naires except those where the response to question 1 was "No."
Question 13. Same as question 12.
When the editing is completed, the questionnaires should be coded. This
merely means writing the number of the response to each question on the appro-
priate line near the right-hand margin of the questionnaire. If there was more
than one response to question 3, the number of each type of pollution mentioned
may be written in the margin.
When the coding is completed, each questionnaire should be numbered on
the first page in the blank following "Respondent No." The same sequence of
numbers may be used for the test area and the control area, since the areas
would be distinguished by the survey numbers assigned earlier .
Tabulation may be accomplished in one of two ways. A form can be de-
signed on which to tally the number of responses of each kind to each question.
(A recommended form is provided at the end of this procedure.) An alternative
is to sort the questionnaires themselves into stacks according to the answers to
a particular question and count the number of questionnaires in each stack. This
method tends to be somewhat more error free than tallying.
Regardless of the method used, the total of all responses to each ques-
-14-
-------
tion (except question 3) must match the total number of people asked that ques-
tion. This provides a check on the accuracy of the tabulation. (Since question
3 is a multiple response question, the total number of all responses may exceed
the total number of people asked that question. Questionnaires containing more
than one response to question 3 should be separated from those having only one
response . Tabulate and total the responses on questionnaires having only one
response. Then, tabulate separately the responses for each kind of pollution
mentioned on the questionnaires with more than one response . Finally, add the
two tabulations together to obtain the total number of responses for each kind of
pollution included in question 3.)
For purposes of odor problem identification, it is not necessary to tabu-
late the responses to all questions. Question 1 is included in the survey to avoid
errors caused by the use of street address telephone directories that may be as
much as six months old. (Such directories are published at six-month intervals.)
Questions 2 and 3 are included for introductory purposes. Questions 4 through
10 provide odor problem definition. Question 11 permits possible identification
of the odor source. Questions 12 and 13 are included so that employees of the
source under investigation may be removed from the sample .
All questions may be tabulated and will yield useful information, but for
odor problem identification, tabulation should begin with question 9. The re-
sponses to question 9 alone will indicate whether an odor problem exists . If it
is determined that an odor problem does not exist, complete tabulation of all
questions need not be undertaken.
PROBLEM IDENTIFICATION
The decisive step in odor problem identification is to determine whether
a significant difference exists between the proportion of respondents in the test
area who state they have been bothered by odors and the proportion of respondents
in the control area who state they have been bothered by odors . This is done by
using the tabulated results of question 9 from the test area and control area sur-
veys . A form entitled "Odor Problem Identification" is included at the end of
this procedure to provide a convenient method for determining whether a signi-
ficant difference exists. A technical explanation of the method is presented in
the "Statistical Reference" section of this procedure.
If, after completing the "Odor Problem Identification" form, the number
in the box labeled "z" is equal to or larger than 1.65, a significant difference
exists between the test and control areas . If "z" is less than 1.65, no signifi-
cant difference exists. If this analysis indicates no significant difference, then
it should be concluded that the test area is not an odor problem area. If a signi-
ficant difference is indicated, then an odor problem index number should be
calculated to indicate the extent of the odor problem relative to other odor
-15-
-------
problems identified within the jurisdiction of the local agency. The odor problem
index number may be simply constructed by completing a form entitled "Determi-
nation of the Odor Problem Index." This form is also included below. It is based
on a formula given in the "Statistical Reference" section.
OTHER USES OF SURVEY RESULTS
During abatement proceedings, it may be necessary to refer to the results
of the test area survey to indicate the percentage of test area residents who re-
sponded in a particular way. For example, it may be necessary to state the
percentage of all test area residents who noticed odors in their neighborhood
during the week of the survey. It would be inaccurate, however, to use directly
the percentage of test area residents who answered "This week" to question 8.
An adjustment is required to eliminate responses to question 8 that are actually
based on variables other than odors .
The adjustment is easy to accomplish. Simply subtract the percentage
response to the question in the control area from the percentage response in the
test area. The answer is then divided by one hundred percent minus the percent-
age response in the control area. The resulting percentage is the true percentage
response for the test area. Thus, if 65% of test area residents and 25% of control
area residents stated they noticed odors in their neighborhoods "This week, " then
53% would represent the true percentage of test area residents who actually noticed
odors in their neighborhood during the week of the survey (65% - 25%/(100% - 25%)
= 53%).
-16-
-------
PUBLIC ATTITUDE SURVEY OF COMMUNITY ODOR PROBLEMS
PROBLEM IDENTIFICATION QUESTIONNAIRE
Respondent:
Phone No.
Street Address
City
File No.
Survey No.
Respondent No.
Record of Call:
Date
Time
Check If
Interview
Completed
Check If
Interview Not
Completed
Reason If
Not Completed
(name)
. I'm calling (long dis-
"Hello. My name is
tance) for a government agency interested in certain community problems.
I'd like to talk with the man/lady of the house to get his/her opinion on a few
questions. Are you the man/lady of the house?" (IF NOT, ASK: "May I talk
with him/her?")
Ql) Have you lived at your present address more than six months?
1. Yes
2. No (SAY: "I'm sorry. This survey was designed for
longer term residents. Thank you for your time.")
Q2) Lately, people have become very much concerned about the environ-
ment and various types of pollution. Do people complain about any
kind of pollution in your neighborhood?
1. Yes (ASK Q3.)
2. No(ASKQ4.)
3. Don't know (ASK Q4.)
Page 1
-------
Q3) Do people complain about...?
1. Air pollution in your neighborhood
2. Water pollution in your neighborhood
3. Airport, industrial, or traffic noise in your neighborhood
4. Noticeable odors in your neighborhood
5. Don't know (DO NOT READ THIS RESPONSE .)
Q4) Have you noticed any odors in your neighborhood in the last three
months?
1. Yes (ASK Q5.)
2. No (SKIP TO Q12.)
3. Don't know (SKIP TO Q12.)
Q5) How often have you noticed these odors?
1. At least once a day
2. At least once a week, but not every day
3. At least once a month, but not every week
4. Less than once a month
5. Don't know
Q6) Generally speaking, how long do these odors last?
1. At least one day
2. At least one hour, but less than one day
3. At least 15 minutes, but less than one hour
4. Less than 15 minutes
5. Don't know
Page 2
-------
Q7) How strong would you say these odors smell? Would you say...?
1. Very strong
2. Strong
3. Moderate
4. Slight
5. Don't know (DO NOT READ THIS RESPONSE .)
Q8) When was the last time you noticed odors in your neighborhood?
1. This week
2. Last week
3. Two to four weeks ago/a month ago
4. More than a month ago
5. Don't know
Q9) Would you say that these odors have bothered you?
1. Yes(ASKQ10.)
2. No (SKIP TO Qll.)
3. Don't know (SKIP TO Ql 1.)
Q10) How much would you say they have bothered you? Would you say...?
1. Very much
2. Much
3. Moderately 10_
4. Little
5. Don't know (DO NOT READ THIS RESPONSE .)
Page 3
-------
Qll) Where would you say most of these odors originate, that is, who or
what causes them?
1. Source(s) being investigated
2. Other source(s):
11
3. Don't know
"All this information is strictly confidential, but we need it for statistical
purposes."
Q12) What company do you work for?
1. Source(s) being investigated
2. Other company/none:
12
3. Refused
Q13) What companies do other members of your household work for?
1. Source(s) being investigated
2. Other company(ies)/none:
13
3. Refused
"That completes the interview. Thank you very much for your time."
Interviewer's Name:
Page 4
-------
PUBLIC ATTITUDE SURVEY - TABULATION FORM
File No.
Survey No.
Instructions. Place a tally mark in the appropriate box to record the answer to each completed
question onthe questionnaire. By tallying the fifth duplicate response with a diagonal line, count-
ing total responses is simplified ( TH4. = 5 responses).
Question 1. (Tabulate only if desired)
Response 1
Total Response 1
Response 2
Total Response 2
Total
Question 2. (Tabulate only if desired)
Response 1
Response 2
Response 3
Total Response 1
Total Response 2
Total Response 3
Total
Question 3. (Tally all responses given on each questionnaire, if desired)
Rl
Total R 1
R2
Total R2
R3
Total R 3
R4
Total R 4
R5
Total R 5
Total
Question 4.
Rl
Total R 1
R2
Total R 2
R3
Total R 3
Total
Question 5.
Rl
Total R 1
R2
Total R2
R3
Total R 3
R4
Total R 4
R5
Total R 5
Total
Paee^L
-------
PUBLIC ATTITUDE SURVEY - TABULATION FORM (Continued)
File No.
Survey No.
Question 6.
Rl
Total R 1
R2
Total R 2
R3
Total R 3
R4
Total R 4
R5
Total R 5
Question 7 .
Rl
Total R 1
R2
Total R 2
R3
Total R 3
R4
Total R 4
R5
Total R 5
Question 8.
Rl
Total R 1
R2
Total R 2
R3
Total R 3
R4
Total R 4
R5
Total R 5
Question 9.
Rl
Total R 1
R2
Total R 2
R3
Total R 3
R4
Total R 4
R5
Total R5
Question 10.
Rl
Total R 1
R2
Total R 2
R3
Total R 3
R4
Total R 4
R5
Total R 5
Total
Total
Total
Total
Total
Page 2
-------
PUBLIC ATTITUDE SURVEY - TABULATION FORM (Continued)
Question 11. (Tabulate only if desired)
Rl
R2
File No.
Survey No.
R3
Total R1
Total R2
Total R 3
Total
Question 12. (Tabulate only if desired)
Rl
R2
R3
Total R1
Total R2
Total R 3
Total
Question 13. (Tabulate only if desired)
Rl
Total R 1
R2
Total R 2
R3
Total R 3
Total
Page 3
-------
ODOR PROBLEM IDENTIFICATION
Instructions
1. In each box labeled "A," fill in the number of interviews completed in the Control Area survey. (The number of interviews
completed = the number of problem identification questionnaires checked as completed.)
2. In each box labeled "B," fill in the number of interviews completed in the Test Area survey.
3. In each box labeled "C," fill in the number of "Yes" responses (R 1) to Question 9 from the Test Area survey.
4. In each box labeled "D," fill in the number of "Yes" responses (R 1) to Question 9 from the Control Area survey.
5. Follow the direction of the arrows. Perform the calculations indicated. (The square root can be easily found by referring
to a standard table of square roots and, therefore, need not be calculated.)
6. If the number in the box labeled "z" is 1.65 or more, an odor problem exists.
Multiply
Subtract
Divide
t
Multiply
A
C
A
C
X
Add
+
Add
+
Add
+
B
D
B
D
Find Square
Root
A
A
Add
-------
DETERMINATION OF THE ODOR PROBLEM INDEX
Instructions
Insert the total numbers of responses tabulated from the surveys in the boxes provided. Follow the direction
of the arrows. Perform the calculations indicated. Round all decimal numbers to the nearest tenth.
Test Area Responses
Rl R2
Q5 "
Q6
Q7
Q8
Q4
Q10
Q9
Control Area Responses
Rl R2
Q5 "
Q6
Q7
Q8
Q4
Q10I|
Q9 I I
CZZ1+
t
-CZI
A
X
1x2.5 =
x I |x2.5 =
A
V
OPI
-------
STATISTICAL REFERENCE SECTION
Statistically Significant Differences for Odor Problem Identification
To determine whether there is a significant difference between the test and
control area responses to question 9 of the Problem Identification Questionnaire,
the following hypotheses must be tested:
Hypothesis 1 (Hi): -nt = IT
*- c
Hypothesis 2 (H2): irt > TC (one -tail test)
a = 0.05
where: irt = the true or universe proportion of "Yes" responses
to question 9 in the test area
ir = the true or universe proportion of "Yes" responses
to question 9 in the control area
a = the probability of concluding, on the basis of sample
values, that no significant difference exists between
test and control areas when a significant difference
in universe values exists
Using the normal approximation:
*t nc
where: i^ = the sample size, i.e., the number of interviews com
pleted in the test area
nc = t^ie samPle size, i.e., the number of interviews com
pleted in the control area
kj. = the number of "Yes "responses to question 9 in the
test area
kc = the number of "Yes" responses to question 9 in the
control area
-26-
-------
pt =
"t
^L
nc
Assuming the null hypothesis is true, this equation becomes:
(2) z =
-------
Determination of the Odor Problem Index
OPI = 2.5
(R1 + R2)
Q10
(Rl)
Q4
(Rl)
Q9
Test Area
- 2.5
(R1 + R2) +(R1 + R2) +
-------
Adjustment to Eliminate Test Area Responses Based on Variables Other Than Odors
The formula is a Bayesian type of correction for chance success:
p'(S|s) = P(S|s) - P(Sln)
lOO-P(Sjn)
where: P'(Sjs) = the percentage of "signal" attributable
to the variable under study
P(S |s) = the total percentage of "noise" plus
"signal" measured in the test group
P(S|n) = the percentage of "noise" measured
in the control group
-29-
-------
PROCEDURE FOR SOURCE VERIFICATION
BASED ON SENSORY TECHNIQUES
INTRODUCTION
This procedure is designed to assist local agencies in source verification
of identified odor problems . It describes the proper use of two sensory tech-
niques, one using a Barnebey-Cheney Model 1-3 Scentometer, the other employ-
ing an odor judgment panel trained to evaluate an odorized environment in terms
of memorized reference standards . As implied below, the choice of techniques
for use in a particular case depends upon the extent of information desired.
An odor judgment panel can be used not only to verify the source and
areal extent of odors, but also to judge odor intensities concurrently at points
throughout a test area. The size of an odor panel depends on the size of the test
area and the transiency of the odors encountered. In general, the larger the
test area or the more transient the odors, the larger must be the odor panel.
In most cases, one panelist per city block would provide adequate coverage .
A scentometer can be used to measure ambient odor intensities while
traveling through the test area. When odors are strong and constant, one scent-
ometer may be adequate to cover a square mile or more. However, when odors
are of low intensity and transient, one scentometer may be adequate to cover
only a few city blocks .
WHEN TO USE THE SCENTOMETER
The scentometer is best suited for use in routine surveillance of chronic
situations. It should be used to determine if odors can be detected above a speci-
fied dilution to threshold and, thereby, to indicate the locations of possible odor
problems.
The scentometer could be used to assist a trained observer to verify the
source of an identified odor problem . Specifically, it could be used:
(1) To establish a pattern of peak odor intensities, which may
point to the source, while systematically traversing the test
area in an automobile.
(2) To function as an odor free air chamber, and so, to refresh
the trained observer whose ability to continuously sense the
odor under investigation may otherwise be lessened due to
fatigue.
-31-
-------
It is noted, however, that although the scentometer is always ready for immedi-
ate use in routine surveillance, it is not always applicable to source verification.
This is because of the difficulty of obtaining readings when the odors encountered
are transient.
WHEN TO USE AN ODOR JUDGMENT PANEL
An odor judgment panel should be used to verify the source of an identi-
fied odor problem if such verification cannot be established by a trained observer
alone or by a trained observer using a scentometer. For source verification, an
odor judgment panel can be used:
(1) To establish patterns of mean odor intensities and percentages
of time odor was detected from concurrent evaluations at
points throughout the test area.
(2) To judge the odor under investigation when odors of different
qualities are encountered in the test area.
Such uses are possible only after lengthy selection and training processes neces-
sary to ensure that the panelists can accurately relate to a series of reference
standards.
THE SCENTOMETER
The scentometer (Figure 1) has been described by its developers and
used by at least twelve local air quality agencies throughout the United States.
Most of the descriptions have focused on the physical characteristics of the de-
vice and the principles of operation. Emphasis in this procedure is based on
experience gained from actual use.
Preparation for Field Use
Before going into the field, the scentometer should be checked for proper
function as follows: Tapes must be placed over the four dilution orifices in such
a way that they can be pulled clear of the orifices and replaced repeatedly. It
should be determined that all odors are absorbed by the charcoal beds. If neces •
sary, fresh charcoal should be installed. Before each use, the scentometer
should be shaken gently in a horizontal position to distribute the charcoal evenly
over the beds, thereby eliminating the possibility of channeling.
The scentometer must be examined for leakage, especially around the
two bulbs which fit to the operator's nostrils. If the bulbs are applied with
pressure to the nostrils to avoid leakage, soreness can develop during constant
-32-
-------
Figure I. Barnebey-Cheney, Model 1-3 Scentomete£,
0
Inches
-33-
-------
use in the field. If the bulbs are placed gently against the nostrils, leakage can
result. Before and during measurements, the operator should check that no
leakage of any type is occurring. This is done by using the scentometer in an
area of strong odor after blocking the four orifices . No odor should be discern-
ible.
Field Measurements
Before field measurements are attempted, the scentometer operator must
first neutralize the effects of ambient odors by inhaling odorless air through the
charcoal beds of the scentometer, while keeping the four orifices blocked. When
this is accomplished, he should open the orifices of the scentometer in turn,
until he can identify the smallest orifice through which odor can be detected.
It is noted that, while this technique is generally successful with strong
and constant odors, very few readings may be obtained with low intensity and
transient odors. Under the latter conditions, it is advantageous to use the
scentometer while driving through a community. The application of this tech-
nique by a mobile operator is fairly simply and considerably more effective.
If odors are encountered in a community during routine surveillance, the
intensities should be measured with the scentometer. Intensities detected at a
particular dilution to threshold would indicate the location of a possible odor
problem. After traversing the community several times, the operator should
relax and breathe odor free air. This can be done by using the scentometer as
an odor free chamber. However, it is more desirable to drive outside the test
area, since it is difficult for the operator to relax with the scentometer bulbs
pressed against the nostrils.
When the operator is suitably refreshed, the car is again driven through
the affected area. Throughout this drive the operator breathes through the
scentometer with a selected orifice open. The route then may be retraced
with a different orifice uncovered, in order to further define the odor intensity.
Each time the route is traversed, the operator calls out the points at which
negative readings change to positive and vice versa. These points should be
plotted on a map of the area or recorded on a form such as the "Scentometer
Log Sheet" provided at the end of this procedure. It is possible to have four
operators and a driver traverse the area in one car. With a different orifice
open on each scentometer, a reading for any odor in the area can be quickly
obtained.
This method, while more effective than scentometer readings taken at a
fixed point, entails some problems. The automobile must be well ventilated
and maneuverable. There must be two to five people in each car. If there are
less than five people in a car, the success of the technique depends upon the con-
-34-
-------
stancy of the odor, since two or more trips through the area would be required
to obtain readings through all of the orifices.
THE ODOR JUDGMENT PANEL
If a local agency wishes to perform odor judgment panel evaluations,
panel members could be recruited from among the agency employees. This
approach could reduce the urgency of preliminary activities and lower costs
considerably, since panelist selection and training could be conducted at lei-
sure at the agency facilities . It must be considered, however, that the results
of evaluations by agency employees could be criticized as being biased against
the source under investigation.
Regardless of this possibility, certain preliminary activities must be
undertaken each time an odor judgment panel is to be used. These activities
include determination of the size of the panel required, location of suitable sta-
tions (points in the test area at which panelists would be placed), location of a
panel headquarters (if the agency facilities cannot be used), and preparation for
transporting the panelists between the headquarters and the test area.
Preliminary Activities
The community in which an odor judgment panel is to be used should be
visited to determine the areal extent of the odors to be investigated. It should
be initially planned to provide one panelist for each city block affected by odors.
If the test area is very large, the size of the panel must be determined by bud-
getary constraints. It is recommended that a minimum of twelve panelists be
used for source verification in any case where public nuisance law is applicable.
The stations should be located in a systematic manner so that panel
members are roughly equally spaced, with intervening distances not exceeding
one city block (approximately 100 yards). Local odor interferences such as
caused by automotive service garages, bakeries, or heavily traveled roadways
should be avoided when selecting stations. Large scale maps of the test area
should be used to plot all stations. Such maps should be sketched if none are
available from commercial or governmental sources.
In addition to selecting stations, convenient locations for performing
panel calibration tests must be found. These locations must be immediately
downwind from an odor source, at places where the presence of large panel
groups will not cause much disturbance in the neighborhood. Public parking lots,
particularly for municipal golf courses and parks, work well.
The headquarters, which is the site used for testing, screening, train-
ing, and assembling, should be located close to the test area. The shorter the
-35-
-------
distance between these locations, the less the expense and proportion of daily
schedule consumed in travel. However, one problem with having a headquarters
close to the site is the possibility of leakage of information to the source under
investigation. Depending upon secrecy requirements, it may be necessary to
locate several miles distance from the site. Once in the test area, the presence
of an odor panel is quickly noticed not only by residents, but possibly by employ-
ees of the facilities under investigation. Therefore, if secrecy is important, it
may be desirable to use the panel for single days only, or for two or more ran-
domly chosen days.
Transportation of the panelists to the test area can be by either cars or
a bus . If a bus is used, all panelists may be transported together, and only one
local agency representative is required to serve as driver. However, a signifi-
cant amount of maneuvering is required in this work, and buses are not readily
maneuverable. Cars are more maneuverable but, as the entire panel cannot be
transported in one car, additional drivers are required. Cars are preferable
as long as the area is known and several drivers are available, while a bus is
preferable to a convoy of cars in a strange area.
To ensure proper selection, training, and deployment of panelists, one
agency representative should be assigned to work with a maximum of six panel-
ists. If such representatives would serve as drivers, the requirement for
additional drivers is easily satisfied. Nine passenger station wagons provide
an ideal compromise between buses and cars . With vehicles of this capacity,
six panelists and their assigned agency representative can be transported
together.
Recruiting Potential Panelists
If sufficient agency employees are not available for this work, it may
be necessary to engage a professional survey firm to recruit panel members .
It is preferable to select panel members with similar socio -economic charac -
teristics to the residents of the test area, as although the panelists are asked
only to relate the odor to standard intensities, bias may be introduced if panel-
ists have a different background. If a panel is required for one week or less,
college students and temporarily unemployed males are available, but if a
panel is needed for longer periods or from time to time at regular intervals,
housewives are usually the most available group.
Initial Selection of Panelists
After sufficient candidates have been recruited, the screening of poten-
tial panelists should be undertaken. Such screening is intended to exclude
individuals with particularly insensitive or irrational responses to odors.
Series of triangle tests are used initially to screen potential panelists. This
method is quick, easy, and inexpensive to perform. The necessary equipment
-36-
-------
consists of a supply of 2-ounce paper cups, water, and odorous food extracts
such as vanilla and wintergreen. The prospective panel members are given
three cups to sniff and are instructed to specify the cup with the different odor.
The three cups presented to the panelists may be one of the three following com-
binations :
(1) Two with the same odor, one with a different odor.
(2) Two with water (no odor), one with an odor.
(3) Two with the same odor, one with water (no odor).
It is best to begin this test using strongly scented samples detectable to
everyone. The concentration of extract should be gradually decreased in water
toward threshold levels . If the tests are conducted over several hours, breaks
should be taken at regular intervals to avoid sensory fatigue. The test room
should be well ventilated and odor free. Using an activated carbon room puri-
fier unit, such as the Barnebey-Cheney Model ABB (4,000 cfm) or Model ADB
(10,000 cfm), may be necessary to ensure an absence of ambient odor.
After each test involving all panelists, the scores should be reviewed to
ascertain the percentage of correct answers. A range of correct answers should
be chosen as a target, say 50% to 75%. If, for most prospective panelists, the
actual percentage is outside this range, the concentration of odorants must be
readjusted. When 6 to 9 tests have been completed, potential panelists should
be accepted or rejected on the basis of their scores. An acceptance range of plus
and minus two correct answers from the mean of the group is recommended.
Final Selection of Panelists
After the preliminary screening by triangle testing, training with refer-
ence standards is begun. A non-toxic, stable standard(s) should be chosen that
closely matches the odor(s) to be encountered in the test area. Using various
concentrations of this standard, the remaining candidates (now called "panelists")
are trained to recognize the strengths of the odor with which they will be con-
cerned . Any one of a variety of standards may be utilized. In a study of solvent
emissions, for example, a methyl chloroform series may be used for degreas-
ing process odors, a turpentine series for thinner odors, or an amyl acetate
series for lacquer solvent odors. A panel recruited to observe rendering plant
odors can be trained with ethanolamines . For refinery odors, reference stan-
dard solutions of tertiary dodecyl mercaptan are suitable .
The reference standard is dissolved in varying concentrations in an odor-
less solvent and numbered according to strengths. The number of strengths in a
series may vary from 4 to 12, depending upon the range of odor intensities expec-
ed to be encountered and the ability of the trained nose to differentiate between
the strengths of adjacent solutions. As exemplified in Table 2, if the number of
-37-
-------
strengths in a series is limited to four, each solution might be four times more
concentrated than the previous solution.
Table 2. Concentration of reference standard in odorless solvent used to
simulate four strengths of the odor expected to be encountered in the test area.
Odor Intensity Concentration in Solvent
Rating (milliliters/liter) Layman Assessment
1 0.125
2 0.5
3 2
4 8
slight
moderate
strong
very strong
The reference standards should be mixed ahead of time, although the un-
diluted odorant and solvent should be brought to the panel headquarters in case
different or additional concentrations must be prepared.
Again, the test room should be well ventilated and odor free . The stan-
dards may be presented in capped, 4-ounce, polyethylene bottles with a polyethy-
lent insert to minimize the possibility of leakage caused by shaking. Even with
these bottles, however, napkins or paper towels should be kept on hand in case
of spills.
Panelists should be instructed to shake the bottle gently once or twice,
remove the cap, and squeeze the sides once as they hold the bottle under the nose.
After a period of memorizing the different strengths and a few breaks to avoid
fatigue, they should be encouraged to test themselves by removing a standard
from the series at random and attempting to judge its intensity. At this time,
the panel instructors should circulate about the room and test the panelists with
unmarked samples. Prior to the training, a few sets of standards should be pre-
pared on which the numbers are not clearly shown but are hidden on labels attached
to the bottom of the bottles . All panelists should be asked to sniff one of these
samples and judge the strength of the odor. Candidates' scores should be record-
ed and combined with the results of field calibration tests (described below) before
the panel selection is finalized.
Instructions to Panelists
When the results of the reference standard tests indicate that the panel-
ists are ready to go into the field, briefing on recording procedure is necessary.
In addition, the following points should be stressed to them:
(1) All evaluation materials, including clipboards, rating sheets,
and pencils, will be supplied by the local agency.
-38-
-------
(2) A watch with a second hand and a large, easily read face
is to be worn by each panelist.
(3) Appropriate clothing, including hats, coats, and gloves,
is to be worn by each panelist if unfavorable weather is
to be expected. (The work involves standing on streets.
Cold wind and dampness, which may be present, can affect
the efficiency of the unprepared panelist.)
(4) Smoking by panelists (and, particularly, by local agency
representatives) is forbidden.
(5) Perfume and other strongly scented toiletries are not to
be worn.
(6) Gum chewing and similar diversions by panelists are dis-
couraged .
(7) The recognition of transient odors requires constant atten-
tion. Panelists must be as alert to the task as possible.
(8) Panelists must rate odors encountered in the test area in
accordance with the strengths of the reference standards
memorized in the test room.
All values should be recorded on an "Odor Intensity Rating Sheet," an
example of which is included at the end of this procedure. How frequently values
should be recorded depends upon the frequency, duration, constancy, and inten-
sity of the odors . Normally, panelists should be asked to record the highest
intensity present during each minute. The time recorded should reflect the end
of each minute. For example: "Time: 10:40 a.m., Rating: 1" means that 1
was the highest intensity perceived between 10:39 a.m. and 10:40 a.m. Similar-
ly, "Time: 3:45 p.m., Rating: 0" means that no odor was detected between
3:44 p .m . and 3:45 p .m.
Determining the "quality" of the odor perceived is often a problem, i.e .,
whether or not an odor encountered is that which bothers test area residents.
Panelists should be instructed as follows:
(1) If the panelist is not sure whether the odor is like that used
during training, he must record the odor intensity as well as
possible and make a notation regarding odor quality.
(2) Only if the panelist is certain that the odor is foreign, i.e.,
if he can positively identify it as coming from some source
-39-
-------
other than that under study) should he record a zero. Even
in such a case, he should describe the quality of the odor
perceived.
Calibration Tests
Uniformity of response among panelists must be monitored during both
the training and field phases of the work. This is done by conducting field cali-
bration tests, i.e., by having all panelists record data at one site . There
should be a distance of at least two yards between panelists to ensure indepen-
dent recordings. Any panelists who consistently records values significantly
different from those of the majority should be rejected. Calibration tests
actually serve two purposes. In addition to indicating the uniformity of the
panelists' responses and, consequently, the reliability of the field values, they
provide additional training and experience in the field.
Work Schedule
To get all panelists to work as a unit and, thereby, avoid those delays
which result in reduced evaluation time in the test area, a rigid adherence to
schedule is required. It has been observed that about one hour is the optimum
time to leave a panelist on station. The panelist's concentration capabilities
deteriorate significantly after working in solitude in one place for an hour. The
type of schedule shown in Table 3 has been found to work well:
Table 3. Recommended work schedule.
Time Period Activity
9:30 a.m.- 9:50 a.m. Arrival of panelists, ready for departure
9:50 a.m.-10:10 a.m. Travel to site, position panelists at first stations
10:10 a.m.-11:20 a.m. Panelists record at first stations
11:20 a.m.-11:40 a.m. Transfer to second stations
11:40 a.m.-12:50 p.m. Panelists record at second stations
12:50 p .m. - 1:10 p .m. Return to headquarters
1:10 p.m.- 2:10 p.m. Lunch, restroom, etc.
2:10 p.m.- 2:20 p .m. Short discussion, ready for departure
2:20 p .m. - 2:40 p .m. Travel to site, position panelists at first stations
2:40 p.m.- 3:50 p .m. Panelists record at first stations
3:50 p.m.- 4:10 p.m. Transfer to second stations
4:10 p.m.- 5:20 p.m. Panelists record at second stations
5:20 p .m. - 5:40 p .m. Return to headquarters
5:40 p.m.- 5:45 p.m. Dismissal of panelists
Total time evaluating odors 4-3/4 hours
Total time working 7-1/4 hours
Total time on break 1 hour
Total time per day 8-1/4 hours
-40-
-------
Panel calibration tests can be incorporated into the schedule by reducing
the 1 hour and 10 minute periods allocated to panelists at individual stations.
Normally, one 10 minute calibration test in the morning and another of equal
duration in the afternoon is sufficient to monitor uniformity of response.
General Observations
Based on the results of the use of odor judgment panels during the national
survey of the odor problem and the study of the social and economic impact of
odors, some observations are offered for local agency consideration.
(1) In general, women seem more emotionally involved than
men and, therefore, appear less detached and specific.
(2) Some panelists become very anxious during calibration
runs, whereas they are relaxed when alone during actual
field measurements.
(3) Some panelists have difficulty in concentrating if left alone
for long periods, and start talking to passers-by, listening
to radios, etc. Some discipline is required.
(4) Panelists should be instructed to try to avoid direct ques -
tions from passers-by and, if necessary, should indicate
that they are surveying cars or other objects unrelated to
odors. Some are able to be convincing without difficulty,
while others may have trouble.
(5) Panelists are sometimes questioned by police. It is there-
fore necessary to inform the local police of the field schedule
before the evaluations take place.
SOURCE VERIFICATION
The data recorded by individual panelists can be summarized by panelist
and by station. (Based on the recommended work schedule, as many as four
panelists per day could be assigned to the same station.) Since the relative ability
of each panelist to the panel as a whole may be estimated by use of the calibration
tests, the most appropriate summary is by stations.
Psychologists generally prefer the use of median values as a measure of
central tendency of data generated by human evaluation. But, because of the
typically skewed distribution of odor intensity ratings by odor judgment panelists
(particularly when odors are transient, many "no odor" ratings are listed), the
mean values recorded at each station are recommended for summary purposes.
-41-
-------
A form entitled "Odor Intensity Rating Summary Sheet" is provided at the conclu-
sion of this procedure for the convenience of local agency personnel in preparing
summary information.
A summary of odor judgment panel data can be used to plot odor inten-
sity isopleths for the test area investigated. Such isopleths may show mean
odor intensities or the percentages of time that odor was detected. Examples
of typical isopleths are shown in Figures 2 and 3. By plotting isopleths, study-
ing recorded wind data for the periods of panel evaluation, and noting the topo-
graphical features of the test area, the odor source usually can be verified with
confidence and a pattern of odor intensity levels throughout the test area can be
established.
-42-
-------
Figure 2. Example of isopleths representing
mean odor intensities recorded by odor judg-
ment panelists stationed in a test area at the
points indicated.
Figure 3. Example of isopleths representing
percentages of time odor was detected by odor
judgment panelists stationed in a test area at
the points indicated.
-------
File No.
Community(s) Affected
Operator 1
Operator 2
SCENTOMETER LOG SHEET
Odor Source(s)
Operator 3
Date
Operator 4
Points at Which
Readings Were Taken
Street
Traveled
Cross Street
or Landmark
Operator No . and D/T Setting
Opr:
D/T:
Opr: lOpr:
D/T: ID/T:
Opr:
D/T:
Time and Reading (+ or - )
T
R
T
R
T
R
T
R
Points at Which
Readings Were Taken
Street
Traveled
Cross Street
or Landmark
Operator No . and D/T Setting
Opr:
D/T:
Opr: lOpr:
D/T: ID/T:
Opr:
D/T:
Time and Reading (+ or -)
T
R
T
R
T
R
T
R
-------
ODOR INTENSITY RATING SHEET
NAME DATE
LOCATION
TIME
RATING
COMMENTS
LOCATION
TIME
RATING
COMMENTS
-------
ODOR INTENSITY RATING SUMMARY SHEET
Panelist
Location
(Station)
Percent Time Detected and Mean Odor Intensity
Date:
AM
%
M
PM
%
M
Date:
AM
%
M
PM
%
M
Date:
AM
%
M
PM
%
M
Date:
AM
%
M
PM
%
M
Date:
AM
%
M
PM
%
M
-------
PROCEDURE FOR PROBLEM ASSESSMENT
INTRODUCTION
Tentative methods of problem assessment are contained in the final re-
port, "A Study of the Social and Economic Impact of Odors, Phase II." The
principal economic effect of odors can be measured in terms of property value
differentials, especially residential property value differentials. A description
of the data required and analysis employed in this aspect of problem assessment
is presented in Chapter VI. An effects determination questionnaire was developed
to determine how residents feel odors have affected themselves, their families,
and their properties . Property values, living patterns, and health symptoms are
covered. A description of the questionnaire and its use is included in Chapter VII.
WHEN TO ATTEMPT PROBLEM ASSESSMENT
The most common effect of odors — annoyance — is readily measured
using the procedure for problem identification. Based on results of the above
mentioned study, economic and other social effects are measurable only in an
extremely chronic situation. Such a situation may be presumed to exist if more
than 50 percent of the test area residents interviewed, using the problem identi-
fication questionnaire, state they have been bothered "Very much" by odors in
their community (question 10, response 1).
-47-
-------
APPENDIX
ESTIMATES OF COSTS INCURRED FROM USING PROCEDURES
Estimates of the costs of services and materials necessary for problem
identification and source verification are listed below by task and cost category.
Where possible, these estimates are based on 1971 prices averaged for the
United States as a whole. Consequently, they may differ somewhat from actual
costs incurred in a particular metropolitan area.
Local agency labor requirements are listed in terms of mandays only.
Daily rates must be applied by agency officials.
Public Attitude Survey Costs
The estimates given in this section represent the costs of conducting iden-
tical public altitude surveys in one test area and one control area.
Estimated Cost
Task/Category Unit Total
Task I: Select Test and Control Areas
Supervisory time 1 manday/area 2 mandays
Clerical time 1 manday/area 2 mandays
Purchase of U. S. Census of
Population and Housing Census
Tract Reports (first survey
only) $5 $5
Purchase of census tract maps
for agency's jurisdiction (first
survey only; necessary only if
maps are not included in U. S.
Census of Population and Hous -
ing Census Tract Reports $15 $15
Task II-A: Select Telephone Numbers for Samples
Clerical time 1 manday/area 2 mandays
Rental of street address telephone
directories (for usual six month
period) $50/directory/area $100
-49-
-------
Estimated Cost
Task/Category
Unit
Total
Task II-B: Conduct Interviews
Supervisory time (first survey
only)
Clerical time:
Training (first survey
only)
Interviewing
Reproduction of questionnaires
Telephone toll charges (if appli-
cable)
1 manday
2 mandays
2 mandays/area
120 questionnaires @
$0.20/question-
naire
60 calls/area @
$0.25/call
Task II (Alternative): Select Sample and Conduct Interviews
Reproduction of questionnaires
Professional survey firm charges:
Select sample
Conduct interviews
Telephone toll charges
(if applicable)
Task III: Edit Questionnaires
Clerical time
Task IV: Code Questionnaires
Clerical time
Task V: Tabulate Responses
Clerical time
Reproduction of tabulation forms
Task VI: Evaluate Survey Results
Supervisory time
Clerical time
120 questionnaires @
$0.20/question-
naire
$50/sample/area
60 interviews/area @
$3.50/interview
60 calls/area @
$0.25/call
0.5 manday/area
0 .5 manday/area
2 mandays/area
120 forms @ $0.15/
form
1 manday
1 manday
1 manday
2 mandays
4 mandays
$24
$30
$24
$100
$420
$30
1 manday
1 manday
4 mandays
$18
1 manday
1 manday
-50-
-------
Estimated Cost
Task/Category Unit Total
Estimated Total Costs (Task II Performed By Agency)
First set of surveys:
Supervisory time 4 mandays
Clerical time 17 mandays
Services and materials $192
Each additional set of surveys:
Supervisory time 3 mandays
Clerical time 15 mandays
Services and materials $172
Estimated Total Costs (Task II Performed By Professional Survey Firm)
First set of surveys: ~~
Supervisory time 3 mandays
Clerical time 9 mandays
Services and materials $612
Each additional set of surveys:
Supervisory time 3 mandays
Clerical time 9 mandays
Services and materials $592
Scentometer Measurement Costs
The following estimates represent the use of a scentometer in routine
surveillance for one day.
Estimated Cost
Task/Category Unit Total
Task I: Prepare for Field Use
Purchase of Barnebey-Cheney
Model 1-3 Scentometer
(first use only) $85 $85
Task II: Conduct Field Measurements
Technician time:
Scentometer operator 1 manday 1 manday
Driver 1 manday 1 manday
Agency vehicle use 200 miles @ $0.12/
mile $24
-51-
-------
Estimated Cost
Task/Category
Unit
Total
Task III: Evaluate Measurements
Supervisory time
0.5 manday
0.5 manday
Estimated Total Costs
First use:
Supervisory time
Technician time
Materials
Each additional use:
Supervisory time
Technician time
Materials
0.5 manday
2 mandays
$109
0 - 5 manday
2 mandays
$24
Odor Judgment Panel Costs
The estimates provided below include the costs of selecting, training, and
using twelve odor judgment panelists for source verification in one test area.
Estimated Cost
Task/Category
Unit
Total
Task I: Preliminary Activities
Supervisory time
Agency vehicle use
Task II: Screen Candidates
Supervisory time
Technician time
Rental of odor free room
Purchase of Barnebey-Cheney
Model ABB Activated Carbon
Room Purifier Unit (first
use only)
Purchase of food extracts, mix-
ing bottles, and cups
Payments to candidates (success-
ful candidates only)
1 manday 1 manday
100 miles @ $0.12/
mile $12
1 manday 1 manday
1 manday 1 manday
$25/day $25
$480 $480
$10 $10
12 candidates @ $20/
candidate $240
-52-
-------
Task/Category
Estimated Cost
Unit
Total
Task III: Train Panelists
Supervisory time
Technician time
Rental of odor free room
Purchase of odorant, solvents,
mixing bottles, plastic bottles
Payments to panelists
Task IV: Perform Field Evaluations
Supervisory time
Technician time
Rental of station wagons
Reproduction of rating forms
Payments to panelists
Task V: Summarize Field Evaluations
Supervisory time
Technician time
1 manday
1 manday
$25/day
$50
12 panelists @ $20/
panelist
1 manday
1 manday
2 station wagons @
$17/day plus
$0.17/mile
(200 miles)
80 forms @ $0.05/
form
12 panelists @ $20/
panelist
1 manday
2 mandays
1 manday
1 manday
$25
$50
$240
1 manday
1 manday
$68
$4
$240
1 manday
2 mandays
Estimated Panel Preparation Costs
First panel:
Supervisory time
Technician time
Services and materials
Each additional panel:
Supervisory time
Technician time
Services and materials
Estimated (Daily) Field Evaluation Costs
Supervisory time
Technician time
Services and materials
3 mandays
2 mandays
$1,082
3 mandays
2 mandays
$602
2 mandays
3 mandays
$312
-53-
-------
SUGGESTED REFERENCES
Census Tract Data
U.S. Bureau of the Census. U.S. Censuses of Population and Housing: 1960.
Census Tracts. Washington, D.C.: U. S. Government Printing
1962.
Census tract data for use by each local agency are recorded in one of
a series of reports under the above title. The Bureau of the Census
should be contacted to determine the report number for the agency's
jurisdiction. Census tract data for 1970 should be available in simi-
lar form in early 1972. Census tract maps are included with these
reports.
Odor Judgment Panel Information
Wittes, Janet, and Turk, Amos. "The Selection of Judges for Odor Discrimina-
tion Panels," Correlation of Subjective-Objective Methods in the Study
of Odors and Taste. Special Technical Publication No. 440, American
Society for Testing and Materials, 1968.
This comprehensive discussion of the use and analysis of screening
tests was co-authored by Amos Turk, the originator of the odor judg-
ment panel technique.
Scentometer Information
Huey, Norman A., Broering, Louis C., Jutze, George A., and Gurber, Charles
W. "Objective Odor Pollution Control Investigations," Journal of the Air
Pollution Control Association, X, No. 6 (December, 1960), 441-446.
This article, by the developers of the scentometer, provides the most
comprehensive description available of the physical characteristics of
the device and the principles of operation.
Statistical Tables
Mathematical Tables From Handbook of Chemistry and Physics . Cleveland,
Ohio: Chemical Rubber Publishing Company, any edition.
Tables of square roots, needed for Odor Problem Identification (see page
24), and other values are included in this mathematical section of the
widely used handbook.
-54-
-------
The Rand Corporation. A Million Random Digits With 100,000 Normal Deviates
Glencoe, Illinois: The Free Press, 1955.
This book contains an extensive listing of 5-digit random numbers.
These numbers can be used to select an unbiased sample of test area
and control area telephone numbers from a street address telephone
directory (see page 7, "Obtaining a Sample"). The first two digits of
any of the 5-digit numbers would provide a starting point for areas
having zero to 99 listed telephones . Similarly, the first three digits
would provide a starting point for areas having zero to 999 listed tele-
phones .
-55-
------- |